Adaptive Control is Reversed Between Hands After Left Hemisphere Stroke and Lost Following Right Hemisphere Stroke

*Rini Varghese, James E Gordon, Robert L Sainburg, Carolee J Winstein, Nicolas Schweighofer*

Now published in the Proceedings of the National Academy of Sciences Analysis based on reviewer comments in NB section #9

library(IRdisplay)
display_html(
'<script>  
code_show=true; 
function code_toggle() {
  if (code_show){
    $(\'div.input\').hide();
  } else {
    $(\'div.input\').show();
  }
  code_show = !code_show
}  
$( document ).ready(code_toggle);
</script>
  <form action="javascript:code_toggle()">
    <input type="submit" value="Click here to toggle on/off the raw code.">
 </form>'
)

Load libraries & set theme

rm(list=ls())
options(warn=-1)
ReqdLibs = c("repr","ggplot2","ggpubr","patchwork","IRdisplay","dplyr","tidyr","lemon","lme4","gghalves","robustlmm",
             "lme4","ggExtra","grid","nlme","sjPlot","sjstats","table1","emmeans","broom","svglite","influence.ME","ggplotify")

invisible(lapply(ReqdLibs, library, character.only = TRUE))

thm = theme(
          strip.text.x=element_text(size=20,face="bold"),
          strip.text.y=element_text(size=20,face="bold"),
          legend.text=element_text(size=16,face="bold"),
          legend.position = "top",
          legend.title=element_text(size=16,face="bold"),
          title =element_text(size=14, face='bold'),
          text = element_text(colour = "black",size=18), 
          plot.title = element_text(colour = "black",size = 22, face = "bold"),
          axis.ticks.length = unit(0.3,"cm"),
          axis.line = element_line(colour = "black",size=0.85),
          axis.ticks = element_line(colour = "black",size=0.85),
          axis.text = element_text(colour = "black",size=24),
          axis.title=element_text(size=25)) 

display_markdown("Libraries loaded. Theme set")
options(warn=0)

Libraries loaded. Theme set

annotate_npc <- function(label, x, y, ...)
{
  ggplot2::annotation_custom(grid::textGrob(
    x = unit(x, "npc"), y = unit(y, "npc"), label = label, ...))
}

Hypothesis & predictions

Null Hypothesis: Responsibility assignment is fully flexible
Alt. Hypothesis: Responsibility assignment relies on right hemisphere specialization

Hand = c(rep("Left Hand", 3), rep("Right Hand", 3))
Group = rep(c("Left Hemiparesis", "Controls", "Right Hemiparesis"), 2)
Pred = c(rep("alt", 6), rep("null", 6))
#     LLHP1 LCTRL1 LRHP1 RLHP1 RCTRL1 RRHP1 LLHP0 LCTRL0 LRHP0 RLHP0 RCTRL0 RRHP0
g = c(0, -0.5, 0, 0, 0, -0.5, -0.75, -0.5, 0, 0, 0, -0.5)
acf = c(0, 0.5, 0, 0, 0, 0.5, 0.75, 0.5, 0, 0, 0, 0.5)


hypoDat = data.frame(Hand, Group, Pred, g, acf)
hypoDat$Hand = factor(hypoDat$Hand, levels = c("Left Hand", "Right Hand"))
hypoDat$Group = factor(hypoDat$Group, levels = c("Left Hemiparesis", "Controls","Right Hemiparesis"))
hypoDat

A data.frame: 12 × 5

Hand	Group	Pred	g	acf
<fct>	<fct>	<chr>	<dbl>	<dbl>
Left Hand	Left Hemiparesis	alt	0.00	0.00
Left Hand	Controls	alt	-0.50	0.50
Left Hand	Right Hemiparesis	alt	0.00	0.00
Right Hand	Left Hemiparesis	alt	0.00	0.00
Right Hand	Controls	alt	0.00	0.00
Right Hand	Right Hemiparesis	alt	-0.50	0.50
Left Hand	Left Hemiparesis	null	-0.75	0.75
Left Hand	Controls	null	-0.50	0.50
Left Hand	Right Hemiparesis	null	0.00	0.00
Right Hand	Left Hemiparesis	null	0.00	0.00
Right Hand	Controls	null	0.00	0.00
Right Hand	Right Hemiparesis	null	-0.50	0.50

A data.frame: 12 × 5

Hand	Group	Pred	g	acf
<fct>	<fct>	<chr>	<dbl>	<dbl>
Left Hand	Left Hemiparesis	alt	0.00	0.00
Left Hand	Controls	alt	-0.50	0.50
Left Hand	Right Hemiparesis	alt	0.00	0.00
Right Hand	Left Hemiparesis	alt	0.00	0.00
Right Hand	Controls	alt	0.00	0.00
Right Hand	Right Hemiparesis	alt	-0.50	0.50
Left Hand	Left Hemiparesis	null	-0.75	0.75
Left Hand	Controls	null	-0.50	0.50
Left Hand	Right Hemiparesis	null	0.00	0.00
Right Hand	Left Hemiparesis	null	0.00	0.00
Right Hand	Controls	null	0.00	0.00
Right Hand	Right Hemiparesis	null	-0.50	0.50

options(repr.plot.width = 12, repr.plot.height = 7)
options(warn = -1)

gr1 = ggplot(data = hypoDat, aes(x = Group, y = g-0.2, shape = Group, col = Hand, lty = Pred)) +
    stat_summary(mapping = NULL, data = NULL, show.legend = FALSE, geom = "pointrange",
        fun.data = mean_se, position = position_dodge(width = 0), size = 1.5, lwd = 3,
        alpha = 0.7) + geom_line(aes(group = interaction(Hand, Pred), col = Hand,
    lty = Pred), size = 2, alpha = 1, show.legend = FALSE, position = position_dodge(width = 0.25),
    arrow = arrow(ends = "both", type = "open")) + geom_hline(yintercept = 0, linetype = "dashed",
    size = 1) + coord_capped_cart(ylim = c(-1.38, 1.38), bottom = "both", left = "both") +
    xlab("") + ylab("gain") + scale_color_manual(values = c("#110176", "#E93324")) +
    scale_fill_manual(values = c("#110176", "#E93324")) + scale_shape_manual(values = c(15,
    16, 17)) + theme_pubclean() + thm + annotate_npc("Loss", x = unit(c(0.35), "npc"),
    y = unit(c(0.4), "npc"), gp = gpar(fontsize = 20, fontface = 4, col = "black")) +
    annotate_npc("Amplify", x = unit(c(0.14), "npc"), y = unit(c(0.25), "npc"), gp = gpar(fontsize = 20,
        fontface = 4, col = "black")) + annotate_npc("Switch", x = unit(c(0.65),
    "npc"), y = unit(c(0.25), "npc"), gp = gpar(fontsize = 20, fontface = 4, col = "black")) +

theme(axis.line.x = element_line(colour = "white"), axis.text.x = element_blank(),
    axis.ticks.x = element_blank(), axis.text.y = element_blank(), axis.ticks.y = element_blank(),
    axis.line.y = element_line(arrow = grid::arrow(length = unit(0.7, "cm"), ends = "both")))

gr2 = ggplot(data = hypoDat, aes(x = Group, y = acf +0.2, shape = Group, col = Hand, linetype = Pred)) +
    stat_summary(mapping = NULL, data = NULL, show.legend = FALSE, geom = "pointrange",
        fun.data = mean_se, position = position_dodge(width = 0), size = 1.5, lwd = 3,
        alpha = 0.7) + geom_line(aes(group = interaction(Hand, Pred), col = Hand,
    linetype = Pred), size = 2, alpha = 1, show.legend = FALSE, position = position_dodge(width = 0.25),
    arrow = arrow(ends = "both", type = "open")) + geom_hline(yintercept = 0, linetype = "dashed",
    size = 1) + coord_capped_cart(ylim = c(-1.38, 1.38), bottom = "both", left = "both") +
    xlab("") + ylab("acf") + scale_color_manual(values = c("#110176", "#E93324")) +
    scale_fill_manual(values = c("#110176", "#E93324")) + scale_shape_manual(values = c(15,
    16, 17)) + annotate_npc("Loss", x = unit(c(0.35), "npc"), y = unit(c(0.6), "npc"),
    gp = gpar(fontsize = 20, fontface = 4, col = "black")) + annotate_npc("Amplify",
    x = unit(c(0.14), "npc"), y = unit(c(0.75), "npc"), gp = gpar(fontsize = 20,
        fontface = 4, col = "black")) + annotate_npc("Switch", x = unit(c(0.65),
    "npc"), y = unit(c(0.77), "npc"), gp = gpar(fontsize = 20, fontface = 4, col = "black")) +
    theme_pubclean() + thm + theme(axis.line.x = element_line(colour = "white"),
    axis.text.x = element_blank(), axis.ticks.x = element_blank(), axis.text.y = element_blank(),
    axis.ticks.y = element_blank(), axis.line.y = element_line(arrow = grid::arrow(length = unit(0.7,
        "cm"), ends = "both")))


gr1 + gr2 

options(warn = 0)

Participant characteristics

demo = read.csv("forTable1_v2.csv", stringsAsFactors = FALSE)
library("table1")

demo$Sex = factor(demo$Sex, levels = c("M", "F"), labels = c("Male", "Female"))
# head(demo)

rndr <- function(x, name, ...) {
    if (!is.numeric(x))
        return(render.categorical.default(x))
    what <- switch(name, Age = "Mean (SD)", EHI = "Median [Min, Max]", Chronicity = "Median [Min, Max]",
        UEFM = "Median [Min, Max]")
    parse.abbrev.render.code(c("", what))(x)
}
strata <- c(split(demo, demo$SOL))
labels <- list(variables = list(Age = "Age (years)", Sex = "Sex", EHI = "Edinburgh Handedness Inventory",
    Chronicity = "Chronicity (months)", UEFM = "UE Fugl-Meyer (/66)"))

tt2 = table1(strata, labels, render = rndr)
display_html(tt2)

	Controls (N=20)	Left Hemiparesis (N=15)	Right Hemiparesis (N=15)
Age (years)
Mean (SD)	26.9 (3.99)	59.4 (14.3)	63.5 (10.0)
Sex
Male	9 (45.0%)	8 (53.3%)	11 (73.3%)
Female	11 (55.0%)	7 (46.7%)	4 (26.7%)
Edinburgh Handedness Inventory
Median [Min, Max]	89.4 [44.0, 100]	100 [50.0, 100]	100 [60.0, 100]
Chronicity (months)
Median [Min, Max]	NA	93.0 [23.0, 211]	95.0 [19.0, 141]
UE Fugl-Meyer (/66)
Median [Min, Max]	NA	52.0 [13.0, 64.0]	55.0 [20.0, 63.0]

Within-trial analysis (g)

Read data

data = read.csv("YNC_AllData_120321_full.csv", stringsAsFactors = FALSE)  #

data %>%
    filter(CondG == c("BmC")) %>%
    mutate(Adj = PVDE - IDE) %>%
{
    bmCData <<- .
}

head(bmCData)

A data.frame: 6 × 33

	Grp	Subj	SubNo	Cond	CondG	Dist	Trial	Hand	CT	RT	⋯	FDIST	CURV	OnAxErr	OffAxErr	XPosPV	YPosPV	XPosOff	YPosOff	XDistPV	Adj
	<chr>	<chr>	<int>	<chr>	<chr>	<int>	<int>	<chr>	<dbl>	<dbl>	⋯	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	YNC	Subj001	1	BmC05	BmC	5	1	left	0.919	-0.03107	⋯	6.28059	0.15171	3.162355	0.1125651	0.9226155	1.041448	0.9231146	1.073100	-0.004901704	-5.20114
2	YNC	Subj001	1	BmC05	BmC	5	2	left	0.982	-0.10269	⋯	4.94105	0.08149	1.562621	1.0774370	0.9346758	1.026062	0.9327633	1.057103	0.007158609	-3.49286
3	YNC	Subj001	1	BmC05	BmC	5	3	left	0.928	-0.03145	⋯	7.03678	0.16236	1.992743	1.1736929	0.9284034	1.032699	0.9337259	1.061404	0.000886155	-5.48651
4	YNC	Subj001	1	BmC05	BmC	5	4	left	0.992	-0.10407	⋯	6.59775	0.28700	2.478371	-1.5180508	0.9109158	1.035750	0.9068085	1.066261	-0.016601443	-4.83403
5	YNC	Subj001	1	BmC05	BmC	5	5	left	0.946	-0.05807	⋯	3.97641	0.29868	-0.291631	-0.2004563	0.9108948	1.021032	0.9199844	1.038560	-0.016622415	-0.18322
6	YNC	Subj001	1	BmC05	BmC	5	6	left	0.994	-0.09745	⋯	5.60293	0.32918	1.634333	-2.5634031	0.9029615	1.028811	0.8963549	1.057820	-0.024555732	-4.46440

Outlier detection & removal

isnt_out_z <- function(x, thres = 3, na.rm = TRUE) {
  abs(x - mean(x, na.rm = na.rm)) <= thres * sd(x, na.rm = na.rm)
}

compute_group_non_outliers <- . %>%
  # Compute per group mean values of columns
  group_by(group) %>%
  summarise_if(is.numeric, mean) %>%
  ungroup() %>%

  # Detect outliers among groups
  mutate_if(is.numeric, isnt_out_z)  %>% 
  # Remove unnecessary columns
  select_if(Negate(is.numeric))

head(bmCData)
colnames(bmCData)[1]="Grp"
bmCData  %>% 
unite(col = "group", Grp,Subj,Cond,Trial) %>% 
compute_group_non_outliers() %>% 
filter(Adj == FALSE | PV==FALSE | IDE==FALSE & PRDE==FALSE | PVDE==FALSE)  %>%  #
{.$group->> outs}
cat(paste(shQuote(outs, type="cmd"), collapse=", "))

dfTrials = dim(bmCData)[1]/3
perc_rem = (length(outs)/dfTrials)*100 # we divide by 3 because the data frame consists of 3 values (for hands & cursor) per trial

display_markdown(paste("<br>**With these outliers, we will be removing",length(outs),"data points out of ",dfTrials,"(",round(perc_rem,3),"%) of the data**"))


#below printout to rem

A data.frame: 6 × 33

	Grp	Subj	SubNo	Cond	CondG	Dist	Trial	Hand	CT	RT	⋯	FDIST	CURV	OnAxErr	OffAxErr	XPosPV	YPosPV	XPosOff	YPosOff	XDistPV	Adj
	<chr>	<chr>	<int>	<chr>	<chr>	<int>	<int>	<chr>	<dbl>	<dbl>	⋯	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	YNC	Subj001	1	BmC05	BmC	5	1	left	0.919	-0.03107	⋯	6.28059	0.15171	3.162355	0.1125651	0.9226155	1.041448	0.9231146	1.073100	-0.004901704	-5.20114
2	YNC	Subj001	1	BmC05	BmC	5	2	left	0.982	-0.10269	⋯	4.94105	0.08149	1.562621	1.0774370	0.9346758	1.026062	0.9327633	1.057103	0.007158609	-3.49286
3	YNC	Subj001	1	BmC05	BmC	5	3	left	0.928	-0.03145	⋯	7.03678	0.16236	1.992743	1.1736929	0.9284034	1.032699	0.9337259	1.061404	0.000886155	-5.48651
4	YNC	Subj001	1	BmC05	BmC	5	4	left	0.992	-0.10407	⋯	6.59775	0.28700	2.478371	-1.5180508	0.9109158	1.035750	0.9068085	1.066261	-0.016601443	-4.83403
5	YNC	Subj001	1	BmC05	BmC	5	5	left	0.946	-0.05807	⋯	3.97641	0.29868	-0.291631	-0.2004563	0.9108948	1.021032	0.9199844	1.038560	-0.016622415	-0.18322
6	YNC	Subj001	1	BmC05	BmC	5	6	left	0.994	-0.09745	⋯	5.60293	0.32918	1.634333	-2.5634031	0.9029615	1.028811	0.8963549	1.057820	-0.024555732	-4.46440

"LHP_Subj003_BmC05_17", "LHP_Subj004_BmC15_6", "LHP_Subj005_BmC15_17", "LHP_Subj007_BmC05_1", "LHP_Subj007_BmC05_4", "LHP_Subj009_BmC05_10", "LHP_Subj009_BmC05_18", "LHP_Subj009_BmC05_5", "LHP_Subj009_BmC05_8", "LHP_Subj009_BmC15_11", "LHP_Subj011_BmC05_10", "LHP_Subj011_BmC05_12", "LHP_Subj011_BmC05_13", "LHP_Subj011_BmC05_15", "LHP_Subj011_BmC05_17", "LHP_Subj011_BmC05_18", "LHP_Subj011_BmC05_19", "LHP_Subj011_BmC05_20", "LHP_Subj011_BmC15_1", "LHP_Subj011_BmC15_11", "LHP_Subj011_BmC15_12", "LHP_Subj011_BmC15_15", "LHP_Subj011_BmC15_16", "LHP_Subj011_BmC15_19", "LHP_Subj011_BmC15_2", "LHP_Subj012_BmC05_9", "LHP_Subj012_BmC15_7", "LHP_Subj013_BmC05_1", "LHP_Subj013_BmC05_10", "LHP_Subj013_BmC05_11", "LHP_Subj013_BmC05_14", "LHP_Subj013_BmC05_15", "LHP_Subj013_BmC05_19", "LHP_Subj013_BmC05_3", "LHP_Subj013_BmC05_5", "LHP_Subj013_BmC05_7", "LHP_Subj013_BmC05_8", "LHP_Subj013_BmC05_9", "LHP_Subj013_BmC15_20", "LHP_Subj014_BmC05_15", "LHP_Subj016_BmC15_8", "RHP_Subj001_BmC05_13", "RHP_Subj001_BmC15_2", "RHP_Subj002_BmC05_7", "RHP_Subj005_BmC15_15", "RHP_Subj006_BmC15_14", "RHP_Subj007_BmC05_2", "RHP_Subj007_BmC15_6", "RHP_Subj008_BmC05_15", "RHP_Subj011_BmC05_3", "RHP_Subj011_BmC05_8", "RHP_Subj013_BmC05_12", "RHP_Subj013_BmC05_13", "RHP_Subj013_BmC05_8", "YNC_Subj004_BmC05_16"

With these outliers, we will be removing 55 data points out of 2000 ( 2.75 %) of the data

# rem2 = c("LHP_Subj002_BmC05_10", "LHP_Subj003_BmC05_17", "LHP_Subj004_BmC05_5", "LHP_Subj004_BmC15_20", "LHP_Subj004_BmC15_6", "LHP_Subj005_BmC15_17", "LHP_Subj007_BmC05_1", "LHP_Subj007_BmC05_4", "LHP_Subj007_BmC15_18", "LHP_Subj008_BmC05_10", "LHP_Subj009_BmC05_10", "LHP_Subj009_BmC05_13", "LHP_Subj009_BmC05_18", "LHP_Subj009_BmC05_4", "LHP_Subj009_BmC05_5", "LHP_Subj009_BmC05_8", "LHP_Subj009_BmC15_10", "LHP_Subj009_BmC15_11", "LHP_Subj009_BmC15_16", "LHP_Subj011_BmC05_10", "LHP_Subj011_BmC05_17", "LHP_Subj011_BmC05_18", "LHP_Subj011_BmC05_19", "LHP_Subj011_BmC05_20", "LHP_Subj011_BmC05_6", "LHP_Subj011_BmC05_9", "LHP_Subj011_BmC15_1", "LHP_Subj011_BmC15_11", "LHP_Subj011_BmC15_12", "LHP_Subj011_BmC15_13", "LHP_Subj011_BmC15_15", "LHP_Subj011_BmC15_16", "LHP_Subj011_BmC15_19", "LHP_Subj011_BmC15_2", "LHP_Subj012_BmC05_6", "LHP_Subj012_BmC05_9", "LHP_Subj012_BmC15_7", "LHP_Subj012_BmC15_8",  "LHP_Subj013_BmC05_10",  "LHP_Subj013_BmC05_9", "LHP_Subj013_BmC05_15", "LHP_Subj013_BmC05_19", "LHP_Subj013_BmC05_3", "LHP_Subj013_BmC05_5", "LHP_Subj013_BmC05_7", "LHP_Subj013_BmC05_8",  "LHP_Subj014_BmC05_15", "LHP_Subj014_BmC05_6", "LHP_Subj016_BmC15_8", "RHP_Subj001_BmC05_13", "RHP_Subj001_BmC15_14", "RHP_Subj001_BmC15_2", "RHP_Subj002_BmC05_7", "RHP_Subj003_BmC05_17", "RHP_Subj003_BmC05_8", "RHP_Subj003_BmC15_8", "RHP_Subj005_BmC15_15", "RHP_Subj006_BmC05_3", "RHP_Subj006_BmC05_8", "RHP_Subj006_BmC15_14", "RHP_Subj006_BmC15_7", "RHP_Subj007_BmC05_2", "RHP_Subj007_BmC15_3", "RHP_Subj007_BmC15_6", "RHP_Subj008_BmC05_15", "RHP_Subj011_BmC05_16", "RHP_Subj011_BmC05_3", "RHP_Subj011_BmC05_8", "RHP_Subj012_BmC15_15", "RHP_Subj012_BmC15_6", "RHP_Subj013_BmC05_12", "RHP_Subj013_BmC05_13", "RHP_Subj013_BmC05_17", "RHP_Subj013_BmC05_8", "RHP_Subj015_BmC05_5", "YNC_Subj004_BmC05_16", "YNC_Subj009_BmC05_4"
# )

rem2 = c("LHP_Subj003_BmC05_17", "LHP_Subj004_BmC15_6", "LHP_Subj005_BmC15_17", "LHP_Subj007_BmC05_1", "LHP_Subj007_BmC05_4", "LHP_Subj009_BmC05_10", "LHP_Subj009_BmC05_18", "LHP_Subj009_BmC05_5", "LHP_Subj009_BmC05_8", "LHP_Subj009_BmC15_11", "LHP_Subj011_BmC05_10", "LHP_Subj011_BmC05_12", "LHP_Subj011_BmC05_13", "LHP_Subj011_BmC05_15", "LHP_Subj011_BmC05_17", "LHP_Subj011_BmC05_18", "LHP_Subj011_BmC05_19", "LHP_Subj011_BmC05_20", "LHP_Subj011_BmC15_1", "LHP_Subj011_BmC15_11", "LHP_Subj011_BmC15_12", "LHP_Subj011_BmC15_15", "LHP_Subj011_BmC15_16", "LHP_Subj011_BmC15_19", "LHP_Subj011_BmC15_2", "LHP_Subj012_BmC05_9", "LHP_Subj012_BmC15_7", "LHP_Subj013_BmC05_1", "LHP_Subj013_BmC05_10", "LHP_Subj013_BmC05_11", "LHP_Subj013_BmC05_14", "LHP_Subj013_BmC05_15", "LHP_Subj013_BmC05_19", "LHP_Subj013_BmC05_3", "LHP_Subj013_BmC05_5", "LHP_Subj013_BmC05_7", "LHP_Subj013_BmC05_8", "LHP_Subj013_BmC05_9", "LHP_Subj013_BmC15_20", "LHP_Subj014_BmC05_15", "LHP_Subj016_BmC15_8", "RHP_Subj001_BmC05_13", "RHP_Subj001_BmC15_2", "RHP_Subj002_BmC05_7", "RHP_Subj005_BmC15_15", "RHP_Subj006_BmC15_14", "RHP_Subj007_BmC05_2", "RHP_Subj007_BmC15_6", "RHP_Subj008_BmC05_15", "RHP_Subj011_BmC05_3", "RHP_Subj011_BmC05_8", "RHP_Subj013_BmC05_12", "RHP_Subj013_BmC05_13", "RHP_Subj013_BmC05_8", "YNC_Subj004_BmC05_16"
)

bmCData  %>% 
unite(col = "group",Grp,Subj,Cond,Trial) %>% 
filter(!group %in% rem2) %>% 
separate(col = group, into = c("Grp","Subj","Cond","Trial"),sep = "_")  %>% 
pivot_wider(id_cols = NULL, names_from= Hand,values_from = c(CT,RT,PPT,TPV,MT,PV,EE,VE,IDE,PRDE,PVDE,FDE,FPE,IDIST,PVDIST,FDIST,
                                             CURV,OnAxErr,OffAxErr,Adj,XPosPV,YPosPV,XPosOff,YPosOff,XDistPV))  %>% 

pivot_longer(cols = c(Adj_left, Adj_right), names_to = "Hand",values_to = "Adjustment") %>% 
select(Grp,Subj,Cond,Trial,SubNo,CondG,Dist,Hand,IDE_cursor,Adjustment)  %>% 

{.->>bmCData2}

bmCData2$Hand <- factor(bmCData2$Hand, levels = c("Adj_left", "Adj_right"), labels = c("Left", "Right")) 

head(bmCData2)
bmCData2$Grp = factor(bmCData2$Grp, levels=c('YNC','RHP','LHP'))

A tibble: 6 × 10

Grp	Subj	Cond	Trial	SubNo	CondG	Dist	Hand	IDE_cursor	Adjustment
<chr>	<chr>	<chr>	<chr>	<int>	<chr>	<int>	<fct>	<dbl>	<dbl>
YNC	Subj001	BmC05	1	1	BmC	5	Left	-7.20574	-5.20114
YNC	Subj001	BmC05	1	1	BmC	5	Right	-7.20574	6.50602
YNC	Subj001	BmC05	2	1	BmC	5	Left	-6.89997	-3.49286
YNC	Subj001	BmC05	2	1	BmC	5	Right	-6.89997	2.20555
YNC	Subj001	BmC05	3	1	BmC	5	Left	-2.80350	-5.48651
YNC	Subj001	BmC05	3	1	BmC	5	Right	-2.80350	6.49121

g Single Represesentative Subjects

options(repr.plot.width = 13.5, repr.plot.height = 4.5)

Ind_subjGL = 
ggplot(bmCData2 %>% filter(SubNo==36), aes(y = IDE_cursor, x = Adjustment,na.rm = TRUE,group = Hand)) +
       geom_point(aes(col = Hand),stroke = 1.5, size = 3.5, shape = 15, show.legend = FALSE,alpha = 0.3) + 
       geom_hline(yintercept = 0,size = 1,col="gray30",lty="dashed") + 
       geom_vline(xintercept = 0,size = 1,col="gray30",lty="dashed") + 
       stat_cor(aes(group = Hand,col=Hand,label =paste(..r.label.., cut(..p.., breaks = c(-Inf, 0.0001, 0.001, 0.01, 0.05, Inf),
                                                                        labels = c("'****'", "'***'", "'**'", "'*'", "'ns'")),
                                                       sep = "~")),label.x = -20, label.y = c(20,14), size=8,show.legend = FALSE) + 
       coord_capped_cart(xlim = c(-20,20),ylim = c(-40,20),bottom = 'both', left = 'both') + #facet_wrap(~Hand,ncol=2) +
       geom_smooth(aes(col=Hand),  method = 'lm', fill=NA,formula = y~x, lwd=3.5,show.legend = FALSE,fullrange=TRUE) +  
       scale_color_manual(values=c("#110176","#E93324")) +
       scale_fill_manual(values=c("#110176","#E93324")) + 
       scale_alpha_manual(values = c(0.7, 0.2)) +
       theme_pubclean() + thm +
       theme(panel.spacing.x=unit(-2, "lines")) + 
       theme(strip.text.x = element_text(size=0)) + theme(strip.background = element_blank(), strip.text = element_blank()) +
       ylab(expression(bold(Delta*theta))) + 
       xlab(expression(bold(theta)[Cursor]))

# ggsave(file='Ind_subjL.svg', plot=Ind_subjGL, width=4.5, height=4.5)

Ind_subjG = 
ggplot(bmCData2 %>% filter(SubNo==14), aes(y = Adjustment, x = IDE_cursor,na.rm = TRUE)) +
       geom_point(aes(col = Hand),stroke = 1.5, alpha = 0.3, size = 3.5, shape = 16, show.legend = FALSE) + 
       geom_hline(yintercept = 0,size = 1,col="gray30",lty="dashed") + 
       geom_vline(xintercept = 0,size = 1,col="gray30",lty="dashed") + 
       stat_cor(aes(group = Hand,col=Hand,label =paste(..r.label.., cut(..p.., breaks = c(-Inf, 0.0001, 0.001, 0.01, 0.05, Inf),
                                                                        labels = c("'***'", "'***'", "'**'", "'*'", "'ns'")),
                                                       sep = "~")),label.x = -20, label.y = c(20,17),size=7.5,show.legend = FALSE) + 
       coord_capped_cart(xlim = c(-20,20),ylim = c(-12,20), bottom = 'both', left = 'both') + #facet_wrap(~Hand,ncol=2) +
       geom_smooth(aes(col=Hand,fill=Hand), fill=NA,method = 'lm', formula = y~x, lwd=3.5,show.legend = FALSE,alpha=0.1,fullrange=TRUE) +  
       scale_color_manual(values=c("#110176","#E93324")) +
       scale_fill_manual(values=c("#110176","#E93324")) + theme_pubclean() + thm +
       theme(panel.spacing.x=unit(-2, "lines")) + 
       theme(strip.text.x = element_text(size=0)) + theme(strip.background = element_blank(), strip.text = element_blank()) +
       ylab(expression(bold(Delta*theta))) + 
       xlab(expression(bold(theta)[Cursor]))

# ggsave(file='Ind_subjG.svg', plot=Ind_subjG, width=4.5, height=4.5)

Ind_subjGR = 
ggplot(bmCData2 %>% filter(SubNo==21), aes(y = IDE_cursor, x =Adjustment ,na.rm = TRUE,group = Hand)) +
       geom_point(aes(col = Hand),stroke = 1.5, size = 3.5, shape = 17, show.legend = FALSE,alpha = 0.3) + 
       geom_hline(yintercept = 0,size = 1,col="gray30",lty="dashed") + 
       geom_vline(xintercept = 0,size = 1,col="gray30",lty="dashed") + 
       stat_cor(aes(group = Hand,col=Hand,label =paste(..r.label.., cut(..p.., breaks = c(-Inf, 0.0001, 0.001, 0.01, 0.05, Inf),
                                                                        labels = c("'***'", "'***'", "'**'", "'*'", "'ns'")),
                                                       sep = "~")),label.x = -1, label.y = c(20,17), size=7.5,show.legend = FALSE) + 
       coord_capped_cart(xlim = c(-20,20),ylim = c(-10,20),bottom = 'both', left = 'both') + #facet_wrap(~Hand,ncol=2) +
       geom_smooth(aes(col=Hand),  method = 'lm', fill=NA,formula = y~x, lwd=3.5,show.legend = FALSE,fullrange=TRUE) +  
       scale_color_manual(values=c("#110176","#E93324")) +
       scale_fill_manual(values=c("#110176","#E93324")) + 
       theme_pubclean() + thm +
       theme(panel.spacing.x=unit(-2, "lines")) + 
       theme(strip.text.x = element_text(size=0)) + theme(strip.background = element_blank(), strip.text = element_blank()) +
       ylab(expression(bold(Delta*theta))) + 
       xlab(expression(bold(theta)[Cursor]))

# ggsave(file='Ind_subjR.svg', plot=Ind_subjGR, width=4.5, height=4.5)

Ind_subjGALL = Ind_subjGL | Ind_subjG | Ind_subjGR

# ggsave(file='Ind_subjGALL.svg', plot=Ind_subjGALL, width=13.5, height=4.5)
Ind_subjGALL

options(dplyr.summarise.inform = FALSE)
options(warn = -1)
bmCData2 %>%
    group_by(Grp) %>%
    summarise(mean_theta_cursor = mean(IDE_cursor, na.rm = TRUE), se_acf = se(IDE_cursor, na.rm = TRUE))
options(dplyr.summarise.inform = TRUE)

A tibble: 3 × 3

Grp	mean_theta_cursor	se_acf
<fct>	<dbl>	<dbl>
YNC	-4.495351	0.1975443
RHP	-1.889187	0.3112893
LHP	-4.586225	0.4026761

Group-wise estimates for g (from fixed effects)

options(warn = -1)
mixedModEstInter = suppressMessages(lmer(data = bmCData2, IDE_cursor ~ Grp + Grp:Hand:Adjustment +
    (1 + Grp:Hand:Adjustment | SubNo)))
coefs <- coef(summary(mixedModEstInter))
data.frame(coefs[, 1:2])
options(warn = 0)

A data.frame: 9 × 2

	Estimate	Std..Error
	<dbl>	<dbl>
(Intercept)	-4.6264323	1.30975520
GrpRHP	2.9358307	1.99869422
GrpLHP	0.4070487	2.00068142
GrpYNC:HandLeft:Adjustment	-0.5474603	0.06922628
GrpRHP:HandLeft:Adjustment	-0.1155544	0.08116438
GrpLHP:HandLeft:Adjustment	-0.1535723	0.06743758
GrpYNC:HandRight:Adjustment	-0.1442372	0.07575751
GrpRHP:HandRight:Adjustment	-0.2488456	0.04458200
GrpLHP:HandRight:Adjustment	-0.1875744	0.10694754

Subject-wise estimates for g (from random effects)

Left = c(coef(mixedModEstInter)$SubNo[1:20,"GrpYNC:HandLeft:Adjustment"],
         coef(mixedModEstInter)$SubNo[21:32,"GrpRHP:HandLeft:Adjustment"],
         coef(mixedModEstInter)$SubNo[44:47,"GrpRHP:HandLeft:Adjustment"],         
         coef(mixedModEstInter)$SubNo[33:43,"GrpLHP:HandLeft:Adjustment"],
         coef(mixedModEstInter)$SubNo[48:50,"GrpLHP:HandLeft:Adjustment"])

Right = c(coef(mixedModEstInter)$SubNo[1:20,"GrpYNC:HandRight:Adjustment"],
         coef(mixedModEstInter)$SubNo[21:32,"GrpRHP:HandRight:Adjustment"],
         coef(mixedModEstInter)$SubNo[44:47,"GrpRHP:HandRight:Adjustment"],         
         coef(mixedModEstInter)$SubNo[33:43,"GrpLHP:HandRight:Adjustment"],
         coef(mixedModEstInter)$SubNo[48:50,"GrpLHP:HandRight:Adjustment"])


Grp = c(matrix('YNC',20),matrix('RHP',15),matrix('LHP',15)) #,

Subj = c(1:length(Left))
Slopes = data.frame(Subj,Grp,Left,Right)

# Pivot data table for analysis & plotting
Slopes  %>% 
pivot_longer(cols = c(Left,Right),names_to = "Hand",values_to = "g")  %>% 
{.->>MMEstGains}
MMEstGains$Grp = factor(MMEstGains$Grp, levels=c('LHP','YNC','RHP'))
head(MMEstGains)

A tibble: 6 × 4

Subj	Grp	Hand	g
<int>	<fct>	<chr>	<dbl>
1	YNC	Left	-0.54379971
1	YNC	Right	-0.06728542
2	YNC	Left	-0.62510505
2	YNC	Right	-0.12980678
3	YNC	Left	-0.37560941
3	YNC	Right	0.14613980

Plotting g (hypothesis & results)

options(repr.plot.width = 11, repr.plot.height = 7)
options(warn = -1)

rs1 = ggplot(data = MMEstGains, aes(x = Grp, y = g, shape = Grp, col = Hand)) + stat_summary(mapping = NULL,
    data = NULL, fun.data = mean_se, alpha = 0.8, geom = "pointrange", position = position_dodge(width = 0.5),
    size = 1.5, lwd = 3, show.legend = FALSE) + geom_half_point_panel(aes(col = Hand),
    size = 3, alpha = 0.3, position = position_dodge(width = 0.5), show.legend = FALSE) +
    geom_line(aes(group = Hand, col = Hand), size = 2, fun.data = mean_se, alpha = 1,
        position = position_dodge(width = 0.5), stat = "summary", show.legend = FALSE) +
    geom_hline(yintercept = 0, linetype = "dashed", size = 1) + coord_capped_cart(ylim = c(-0.75,
    0.75), bottom = "both", left = "both") + xlab("") + ylab("g") + annotate_npc("paretic left",
    x = unit(c(0.08), "npc"), y = unit(c(0.3), "npc"), rot = 90, gp = gpar(fontsize = 20,
        fontface = 3, col = "#110176")) + annotate_npc("paretic right", x = unit(c(0.85),
    "npc"), y = unit(c(0.2), "npc"), rot = 90, gp = gpar(fontsize = 20, fontface = 3,
    col = "#E93324")) + scale_color_manual(values = c("#110176", "#E93324")) + scale_fill_manual(values = c("#110176",
    "#E93324")) + scale_shape_manual(values = c(15, 16, 17)) + theme_pubclean() +
    thm + theme(axis.line.x = element_line(colour = "white"), axis.text.x = element_blank(),
    axis.ticks.x = element_blank(), legend.position = "none")


Fig1 = gr1 + rs1

# ggsave(file = "Fig1_g.svg", plot = Fig1, width = 11, height = 7)

Fig1
options(warn = 0)

Simple LMER for g & Model Diagnostics

gMod = lmer(g ~ 1 + Hand + Grp + Hand*Grp + (1| Subj), MMEstGains)
estex = influence(gMod,"Subj")

options(repr.plot.width = 14, repr.plot.height = 7)

diagGp1 = as.ggplot(~plot(estex, which="cook",cutoff=.04, sort=TRUE,xlab="Cook's Distance",ylab = "Subj"))
diagGp2 = as.ggplot(~plot(estex, which="dfbetas",cutoff=.28, xlab="DFBetas",ylab = "Subj"))

diagGp1|diagGp2

estex.obs <- influence(gMod, obs=TRUE) 
cks.d <- cooks.distance(estex.obs, parameter=3) 
outliers = which(cks.d > 4/100)

gMod2 = exclude.influence(gMod,"Subj",outliers)

Robust LMER to compare g between groups and hands

Comparing g between YNC & LHP

# SET REFERENCE
MMEstGains$Hand <- factor(MMEstGains$Hand, ordered = FALSE)
MMEstGains$Hand <- relevel(MMEstGains$Hand, "Right")
MMEstGains$Grp <- relevel(MMEstGains$Grp, "LHP")

# RUN ROBUST MODEL
robust.model1L = rlmer(g ~ Hand + Grp + Hand:Grp + (1 | Subj), MMEstGains %>% filter(Grp != "RHP"), doFit = TRUE)

#EMMEANS
summary(emmeans(robust.model1L, specs = pairwise ~ Hand | Grp)$contrasts)
summary(emmeans(robust.model1L, specs = pairwise ~ Grp | Hand)$contrasts)

A summary_emm: 2 × 7

	contrast	Grp	estimate	SE	df	z.ratio	p.value
	<fct>	<fct>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	Right - Left	LHP	-0.02583499	0.03128241	Inf	-0.8258633	4.088816e-01
2	Right - Left	YNC	0.40218868	0.02709136	Inf	14.8456442	7.423821e-50

A summary_emm: 2 × 7

	contrast	Hand	estimate	SE	df	z.ratio	p.value
	<fct>	<fct>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	LHP - YNC	Right	-0.03229614	0.05213391	Inf	-0.6194844	5.355973e-01
2	LHP - YNC	Left	0.39572753	0.05213391	Inf	7.5905983	3.184317e-14

Comparing g between YNC & RHP

# SET REFERENCE
MMEstGains$Hand <- factor(MMEstGains$Hand, ordered = FALSE)
MMEstGains$Hand <- relevel(MMEstGains$Hand, "Right")
MMEstGains$Grp <- relevel(MMEstGains$Grp, "RHP")

# RUN ROBUST MODEL
robust.model1R = rlmer(g ~ Hand + Grp + Hand:Grp + (1 | Subj), MMEstGains %>% filter(Grp != "LHP"), doFit = TRUE)

#EMMEANS
summary(emmeans(robust.model1R, specs = pairwise ~ Hand | Grp)$contrasts)
summary(emmeans(robust.model1R, specs = pairwise ~ Grp | Hand)$contrasts)

A summary_emm: 2 × 7

	contrast	Grp	estimate	SE	df	z.ratio	p.value
	<fct>	<fct>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	Right - Left	RHP	-0.1649034	0.04858103	Inf	-3.394398	6.877959e-04
2	Right - Left	YNC	0.3875822	0.04207241	Inf	9.212266	3.193107e-20

A summary_emm: 2 × 7

	contrast	Hand	estimate	SE	df	z.ratio	p.value
	<fct>	<fct>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	RHP - YNC	Right	-0.09119309	0.0454434	Inf	-2.00674	4.477734e-02
2	RHP - YNC	Left	0.46129252	0.0454434	Inf	10.15093	3.282355e-24

Trial-to-trial analysis (acf)

Read Data

colnames(data)[1] = "Grp"
data  %>% 
filter(CondG==c("BmC")) %>% 
filter(Hand!="cursor") %>% 
mutate(Trial = as.numeric(Trial)) %>% 
select(Grp,Subj,SubNo,Cond,Dist,Hand,Trial,PVDE,XPosPV,YPosPV) %>% 

unite(col = "group",Grp,Subj,SubNo,Cond,Dist,Hand) %>% 
select(group,Trial,PVDE,XPosPV,YPosPV)   %>% 
{.->> ctrlDataF}

head(ctrlDataF)

A data.frame: 6 × 5

	group	Trial	PVDE	XPosPV	YPosPV
	<chr>	<dbl>	<dbl>	<dbl>	<dbl>
1	YNC_Subj001_1_BmC05_5_left	1	-10.32452	0.9226155	1.041448
2	YNC_Subj001_1_BmC05_5_left	2	-6.41764	0.9346758	1.026062
3	YNC_Subj001_1_BmC05_5_left	3	-10.41785	0.9284034	1.032699
4	YNC_Subj001_1_BmC05_5_left	4	-19.34386	0.9109158	1.035750
5	YNC_Subj001_1_BmC05_5_left	5	-9.36932	0.9108948	1.021032
6	YNC_Subj001_1_BmC05_5_left	6	-22.25910	0.9029615	1.028811

Computing Mahalanobis distance & lagged PVDE to determine fast learning

names = unique(ctrlDataF$group)

mahals = vector()
datLags = vector()

for (i in names)

    {dat = ctrlDataF[ctrlDataF$group == i,]
     
     datLag = lag(dat$PVDE,1)
     datLags = c(datLags,datLag)
     
     mahal = mahalanobis(dat[,4:5], 
                    colMeans(dat[,4:5]),
                    cov(dat[,4:5]))
     mahals = c(mahals,mahal)
    }

lagPVDE = datLags

ctrlDataNew = data.frame(ctrlDataF,lagPVDE,mahals)

ctrlDataNew  %>% 
separate(col = group, into = c("Grp","Subj","SubNo","Cond","Dist","Hand"),sep = "_")   %>% 
{.->>ctrlDataNew4reg}

ctrlDataNew4reg$Grp = factor(ctrlDataNew4reg$Grp, levels=c('YNC','RHP','LHP'))
head(ctrlDataNew4reg)

A data.frame: 6 × 12

	Grp	Subj	SubNo	Cond	Dist	Hand	Trial	PVDE	XPosPV	YPosPV	lagPVDE	mahals
	<fct>	<chr>	<chr>	<chr>	<chr>	<chr>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	YNC	Subj001	1	BmC05	5	left	1	-10.32452	0.9226155	1.041448	NA	6.186850
2	YNC	Subj001	1	BmC05	5	left	2	-6.41764	0.9346758	1.026062	-10.32452	4.462720
3	YNC	Subj001	1	BmC05	5	left	3	-10.41785	0.9284034	1.032699	-6.41764	3.160088
4	YNC	Subj001	1	BmC05	5	left	4	-19.34386	0.9109158	1.035750	-10.41785	2.501502
5	YNC	Subj001	1	BmC05	5	left	5	-9.36932	0.9108948	1.021032	-19.34386	1.138511
6	YNC	Subj001	1	BmC05	5	left	6	-22.25910	0.9029615	1.028811	-9.36932	2.812689

Select PVDE for ACF analysis

ctrlDataNew4reg  %>% 
filter(Trial!=1:5) %>%
select(SubNo,Trial,Dist,Hand,PVDE) %>% 
pivot_wider(id_cols=NULL,names_from = Hand,values_from = PVDE) %>% 
{.->>ctrlDataNew4ACF}

head(ctrlDataNew4ACF)

A tibble: 6 × 5

SubNo	Trial	Dist	left	right
<chr>	<dbl>	<chr>	<dbl>	<dbl>
1	6	5	-22.25910	2.23242
1	7	5	-17.25457	8.54909
1	8	5	-18.71507	4.56551
1	9	5	-23.68392	9.18148
1	10	5	-19.89913	16.52577
1	11	5	-7.35605	8.36514

All subjects ACFs

acfL = vector()
acfR = vector()

SubjID = vector()

for (i in ctrlDataNew4ACF$SubNo) {
    acfL[i] = acf(ctrlDataNew4ACF$left[ctrlDataNew4ACF$SubNo == i], lag.max = 1,
        plot = FALSE)$acf[2]
    acfR[i] = acf(ctrlDataNew4ACF$right[ctrlDataNew4ACF$SubNo == i], lag.max = 1,
        plot = FALSE)$acf[2]
    SubjID[i] = i
}


acf = data.frame(acfL, acfR)

colnames(acf)[1] <- "Left"
colnames(acf)[2] <- "Right"
# colnames(acf)[3] <- 'Cursor'


j = c(matrix("YNC", 20), matrix("RHP", 15), matrix("LHP", 15))  #,

acfData = data.frame(SubjID, data.frame(j), acf)

colnames(acfData)[2] = "Grp"

acfData %>%
    pivot_longer(cols = c("Left", "Right"), "Hand", values_to = "acf_coeff") %>%

{
    acfData2 <<- .
}

acfData2$Grp = factor(acfData2$Grp, levels = c("LHP", "YNC", "RHP"))

acf Single Represesentative Subject

eg = 1;
remT = c(1:5)

ctrlDataNew4reg %>% 
select(SubNo,Trial,Dist,Hand,PVDE,lagPVDE) %>% 
filter(SubNo==eg) %>% 

{.->> sub1Dat}
head(sub1Dat,3)


lefteg = acf(ctrlDataNew4ACF$left[ctrlDataNew4ACF$SubNo == eg], plot = FALSE)
righteg = acf(ctrlDataNew4ACF$right[ctrlDataNew4ACF$SubNo == eg], plot = FALSE)
subjACF = data.frame(lefteg$lag, lefteg$acf, righteg$acf)


colnames(subjACF)[1] <- "Lag"
colnames(subjACF)[2] <- "Left"
colnames(subjACF)[3] <- "Right"

subjACF %>%
    pivot_longer(cols = c(Left, Right), names_to = "Hand", values_to = "acf") %>% 
    filter(Lag<8) %>% 
    {subjACF <<- .}

A data.frame: 3 × 6

	SubNo	Trial	Dist	Hand	PVDE	lagPVDE
	<chr>	<dbl>	<chr>	<chr>	<dbl>	<dbl>
1	1	1	5	left	-10.32452	NA
2	1	2	5	left	-6.41764	-10.32452
3	1	3	5	left	-10.41785	-6.41764

options(repr.plot.width = 10, repr.plot.height = 5)


#--------------------ACF PLOT-----------------------#


IndFig = ggplot(data = subjACF, aes(x = Lag, y = acf, fill = Hand)) + 
geom_bar(width = 0.9, position = position_dodge(width = 0.8), stat = "identity", show.legend = FALSE,
    alpha = 0.3) + geom_bar(data = subset(subjACF, Lag == 1), aes(x = Lag, y = acf,
    fill = Hand), width = 1, position = position_dodge(width = 0.9), stat = "identity",
    show.legend = FALSE, alpha = 1) + annotate("segment", x = 0.8, xend = 0.8, y = 0.5, yend = 0.95,
           colour = "#110176", size = 2.2, arrow = arrow()) + #type = "closed",angle = 20,ends = "last",length = unit(0.06, "npc")
annotate("segment", x = 1.2, xend = 1.2, y = -0.09, yend = -0.25,
           colour = "#E93324", size = 2.2, arrow = arrow()) +
annotate_npc("lag = 1", x = unit(c(0.28), "npc"), y = unit(c(0.5), "npc"), rot = 30, gp = gpar(fontsize = 20, fontface = 3,
        col = "black")) +
coord_capped_cart(ylim = c(-0.35, 1.2), xlim = c(0, 8), bottom = "both", left = "both") + 
    xlab("lag") + ylab("acf") + theme_pubclean() + thm + scale_color_manual(values = c("#110176",
    "#E93324")) + scale_fill_manual(values = c("#110176", "#E93324"))
# ggsave(file='IndFig_acf.svg', plot=IndFig, width=5, height=5)


#-------------------- t by (t-1) plot-----------------------#


IndFig2 = ggplot(sub1Dat %>%filter(!Trial%in%remT), aes(y = PVDE, x = lagPVDE,na.rm = TRUE)) +
       geom_point(aes(col = Hand,alpha = Trial),stroke = 0.5, size = 3.5, shape = 16, show.legend = FALSE) + 
       geom_hline(yintercept = 0,size = 1,col="gray30") + 
       stat_cor(aes(group = Hand,col=Hand,label =paste(..r.label.., cut(..p.., breaks = c(-Inf, 0.0001, 0.001, 0.01, 0.05, Inf),
                                                                        labels = c("'****'", "'***'", "'**'", "'*'", "'ns'")),
                                                       sep = "~")),label.x = 0, label.y = c(-21,-17), size=7.4,show.legend = FALSE) + 
       coord_capped_cart(xlim = c(-25,25),bottom = 'both', left = 'both') + 
       geom_smooth(aes(col=Hand,fill=Hand),  method = 'lm', formula = y~x, lwd=2.5,show.legend = FALSE,alpha=0.1) +  
       scale_color_manual(values=c("#110176","#E93324")) +
       scale_alpha("Trial",range = c(0,1),breaks = c(1:20))+
       scale_fill_manual(values=c("#110176","#E93324")) + theme_pubclean() + thm +
       xlab(expression(bold(theta) [italic(peak~vel)~(t-1)])) + 
       ylab(expression(bold(theta) [italic(peak~vel)~(t)]))

# ggsave(file="IndFig2_tbyt-1.svg", plot=IndFig2, width=5, height=5)



IndFig | IndFig2

Mean ± SE for ACFs per group

options(dplyr.summarise.inform = FALSE)
options(warn = -1)
acfData2 %>%
    group_by(Grp,Hand) %>%
    summarise(mean_acf = mean(acf_coeff, na.rm = TRUE), se_acf = se(acf_coeff, na.rm = TRUE))

options(dplyr.summarise.inform = TRUE)
options(warn = 0)

A grouped_df: 6 × 4

Grp	Hand	mean_acf	se_acf
<fct>	<chr>	<dbl>	<dbl>
LHP	Left	0.11830890	0.04667883
LHP	Right	0.08282471	0.06191646
YNC	Left	0.25282072	0.04616468
YNC	Right	0.04451317	0.04740744
RHP	Left	-0.01402382	0.05812553
RHP	Right	0.21097437	0.04067150

Plotting acf (hypothesis & results)

options(repr.plot.width=11, repr.plot.height=7)

options(warn=-1)

rs2 = ggplot(data = acfData2,aes(x = Grp, y = acf_coeff,shape=Grp,col=Hand)) +
      stat_summary(mapping = NULL, data = NULL,show.legend = FALSE,fun.data = mean_se, alpha = 0.8,geom = "pointrange", position = position_dodge(width=0.5),size=1.5,lwd=3) +
      geom_half_point_panel(aes(col=Hand),position = position_dodge(width=0.5),size=3, alpha = 0.2,show.legend = FALSE) +
      geom_line(aes(group=Hand,col=Hand),size=2,alpha = 1,stat = "summary",
                fun.data = mean_se, position = position_dodge(width=0.5),show.legend = FALSE) + 
      geom_hline(yintercept = 0,linetype = "dashed",size = 1) + 
      coord_capped_cart(ylim = c(-0.6,0.6), bottom = 'both', left = 'both') +
      xlab("") + ylab("acf")+ 
      scale_color_manual(values=c("#110176","#E93324")) + 
      scale_fill_manual(values=c("#110176","#E93324")) + 
      scale_shape_manual(values = c(15,16,17)) + 
      annotate_npc('paretic left', x = unit(c(0.08), "npc"), y = unit(c(0.67), "npc"),rot = 90, gp = gpar(fontsize = 20, fontface = 3,col = "#110176")) +
      annotate_npc('paretic right', x = unit(c(0.93), "npc"), y = unit(c(0.73), "npc"),rot = 90, gp = gpar(fontsize = 20, fontface = 3,col = "#E93324")) +
      theme_pubclean() + thm +
      theme(axis.line.x = element_line(colour = "white"),
            axis.text.x=element_blank(),
            axis.ticks.x=element_blank(),
            legend.position = "None")

Fig2m = gr2 + rs2

# ggsave(file="Fig2_acf.svg", plot=Fig2m, width=11, height=7)

Fig2m

Simple LMER for acf & Model Diagnostics

#LMER Model
acfMod = lmer(acf_coeff ~ 1 + Hand + Grp + Hand:Grp + (1| SubjID), acfData2)

# INFLUENTIAL DATA
estex = influence(acfMod,"SubjID")
estex.obs <- influence(acfMod, obs=TRUE) 
cks.d <- cooks.distance(estex.obs, parameter=3) 
outliers = which(cks.d > (4/70))


#PLOT
options(repr.plot.width = 14, repr.plot.height = 7)
diagACFp1 = as.ggplot(~plot(estex, which="cook",cutoff=.06, sort=TRUE,xlab="Cook's Distance",ylab = "Subj"))
diagACFp2 = as.ggplot(~plot(estex, which="dfbetas",cutoff=.28, xlab="DFBetas",ylab = "Subj"))

diagACFp1|diagACFp2

options(repr.plot.width = 15, repr.plot.height = 5.5)

dg1 = as.ggplot(plot(acfMod))
dg2 = as.ggplot(~qqnorm(residuals(acfMod)))
dg3 = as.ggplot(~hist(residuals(acfMod)))

dg1 | dg2 | dg3
shapiro.test(residuals(acfMod))

	Shapiro-Wilk normality test

data:  residuals(acfMod)
W = 0.97471, p-value = 0.05119

acfMod2 = exclude.influence(acfMod,"SubjID",outliers)
# summary(acfMod2)

Robust LMER to compare acf between groups and hands

Comparing acf between YNC & LHP

# SET REFERENCE
acfData2$Hand <- factor(acfData2$Hand, ordered = FALSE)
acfData2$Hand <- relevel(acfData2$Hand, "Right")
acfData2$Grp <- relevel(acfData2$Grp, "LHP")

# RUN ROBUST MODEL
robust.model2L = rlmer(acf_coeff ~ Hand + Grp + Hand:Grp + (1 | SubjID), acfData2 %>%
    filter(Grp != "RHP"), doFit = TRUE)

#EMMEANS
summary(emmeans(robust.model2L, specs = pairwise ~ Hand | Grp)$contrasts)
summary(emmeans(robust.model2L, specs = pairwise ~ Grp | Hand)$contrasts)

A summary_emm: 2 × 7

	contrast	Grp	estimate	SE	df	z.ratio	p.value
	<fct>	<fct>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	Right - Left	LHP	-0.04390767	0.07320295	Inf	-0.5998073	0.5486346439
2	Right - Left	YNC	-0.22578807	0.06339561	Inf	-3.5615725	0.0003686403

A summary_emm: 2 × 7

	contrast	Hand	estimate	SE	df	z.ratio	p.value
	<fct>	<fct>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	LHP - YNC	Right	0.01523264	0.06847509	Inf	0.2224552	0.82395956
2	LHP - YNC	Left	-0.16664777	0.06847509	Inf	-2.4336992	0.01494541

Comparing acf between YNC & RHP

# SET REFERENCE
acfData2$Hand <- factor(acfData2$Hand, ordered = FALSE)
acfData2$Hand <- relevel(acfData2$Hand, "Right")
acfData2$Grp <- relevel(acfData2$Grp, "RHP")


# RUN ROBUST MODEL
robust.model2R = rlmer(acf_coeff ~ Hand + Grp + Hand:Grp + (1 | SubjID), acfData2 %>%
    filter(Grp != "LHP"), doFit = TRUE)

#EMMEANS
summary(emmeans(robust.model2R, specs = pairwise ~ Hand | Grp)$contrasts)
summary(emmeans(robust.model2R, specs = pairwise ~ Grp | Hand)$contrasts)

A summary_emm: 2 × 7

	contrast	Grp	estimate	SE	df	z.ratio	p.value
	<fct>	<fct>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	Right - Left	RHP	0.2557735	0.07037451	Inf	3.634462	0.0002785613
2	Right - Left	YNC	-0.2264390	0.06094612	Inf	-3.715397	0.0002028846

A summary_emm: 2 × 7

	contrast	Hand	estimate	SE	df	z.ratio	p.value
	<fct>	<fct>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	RHP - YNC	Right	0.1611357	0.06582933	Inf	2.447779	1.437397e-02
2	RHP - YNC	Left	-0.3210769	0.06582933	Inf	-4.877414	1.074859e-06

Relationship between trial-by-trial adaptive response and chronicity in RHP

fmchron2 = read.csv("fmchron_subj.csv", stringsAsFactors = FALSE)
acfData %>%
    filter(Grp != "YNC") %>%
    data.frame(., fmchron2[-22, ]) %>%
    filter(Grp == "RHP") %>%
    pivot_longer(cols = c(Left, Right), names_to = "Hand", values_to = "acf") %>%
    {
        forPlots2 <<- .
    }

colnames(forPlots2)[3] = "chron"
head(forPlots2)

A tibble: 6 × 7

SubjID	Grp	chron	uefm	ehi	Hand	acf
<chr>	<chr>	<int>	<int>	<dbl>	<chr>	<dbl>
21	RHP	141	59	100	Left	-0.27089352
21	RHP	141	59	100	Right	0.03707393
22	RHP	95	47	90	Left	-0.10022453
22	RHP	95	47	90	Right	0.13830222
23	RHP	63	51	100	Left	0.04269213
23	RHP	63	51	100	Right	0.27618591

all = data.frame(acfData2,MMEstGains)
# all %>% filter(Grp!="LHP") %>% {.->>all2}
# head(all)
all$Hand = factor(all$Hand, levels=c('Left','Right'),labels = c("Left Hand","Right Hand"))

options(repr.plot.width=12, repr.plot.height=6.5)

Fig3a = ggplot(forPlots2, aes(y = acf, x = chron)) +
       geom_point(aes(col = Hand),stroke = 1.5, alpha = 0.9, size = 3.5, shape = 17, show.legend = FALSE) + 
       geom_hline(yintercept = 0,size = 1,col="gray30",lty="dashed") + 
       coord_capped_cart(xlim = c(0,200),ylim = c(-0.5,0.52),bottom = 'both', left = 'both') + 
       geom_smooth(aes(col=Hand,fill=Hand),  method = 'lm', formula = y~x, lwd=2,show.legend = FALSE,alpha=0.1) +  
       scale_color_manual(values=c("#110176","#E93324")) +
       scale_fill_manual(values=c("#110176","#E93324")) + theme_pubclean() + thm +
       annotate_npc("paretic\n(right hand)", x = unit(c(0.85), "npc"), 
                    y = unit(c(0.61), "npc"), 
                    gp = gpar(fontsize = 18, fontface = 3,col = "#E93324")) +
       annotate_npc("non-paretic\n(left hand)", x = unit(c(0.85), "npc"), 
                    y = unit(c(0.18), "npc"),
                    gp = gpar(fontsize = 18, fontface = 3,col = "#110176")) +
       xlab("chronicity") + ylab("acf")

Fig3b = ggplot(all %>% filter(Grp!="LHP" & Hand=="Left Hand"), aes(x = acf_coeff, y = g)) +
       geom_point(aes(shape = Grp),stroke = 1.5, col="#110176",alpha = 1, size = 3.5, show.legend = FALSE) + 
       geom_hline(yintercept = 0,size = 1,col="gray30",lty="dashed") + 
       geom_vline(xintercept = 0,size = 1,col="gray30",lty="dashed") + 
       geom_smooth(col="#110176",fill="#110176", method = 'lm', formula = y~x, lwd=2,show.legend = FALSE,alpha=0.1) +  
#        facet_grid(~Hand) +  #stat_ellipse(aes(group=Grp)) +
# stat_cor(method="spearman",size = 7) +
       coord_capped_cart(ylim = c(-0.8,0.2),xlim = c(-0.57,0.57), bottom = 'both', left = 'both') + 
      scale_shape_manual(values = c(16,17)) + 
       scale_color_manual(values=c("#110176","#E93324")) +
       scale_fill_manual(values=c("#110176","#E93324")) + 
   theme_pubclean() + thm +
       ylab("gain") + xlab("acf")

Fig4 = Fig3a + Fig3b + plot_annotation(tag_levels = 'A') & theme(plot.tag = element_text(size = 22))
Fig4
# ggsave(file="Fig4.svg", plot=Fig4, width=12, height=6.5)

chron_LRN = lmer(acf ~ chron:Hand + (1 | SubjID), data = forPlots2)
table4 = tab_model(chron_LRN, show.stat = TRUE)
display_html(head(table4$page.complete))

	acf
Predictors	Estimates	CI	Statistic	p
(Intercept)	0.20	0.03 – 0.37	2.33	0.020
chron * HandLeft	-0.00	-0.00 – -0.00	-2.72	0.007
chron * HandRight	0.00	-0.00 – 0.00	0.19	0.849
Random Effects
σ2	0.02
τ00 SubjID	0.01
ICC	0.38
N SubjID	15
Observations	30
Marginal R2 / Conditional R2	0.385 / 0.617

Cursor error comparisons between groups

data  %>% 
filter(CondG==c("BmC")) %>% 
filter(Hand=="cursor")  %>%
select(Grp,Subj,CondG,SubNo,Cond,Dist,Hand,Trial,IDE,PVDE) %>% 
group_by(Subj,CondG) %>% 
mutate(meanIDE = mean(IDE),meanPVDE = mean(PVDE)) %>% 
distinct(Subj,Grp,CondG,meanIDE,meanPVDE) %>% 
{.->>cursorErrPlot}
cursorErrPlot$Grp = factor(cursorErrPlot$Grp, levels = c("LHP", "YNC", "RHP"))
cursorErrPlot$CondG <- as.factor(cursorErrPlot$CondG)
cursorErrPlot$CondG = factor(cursorErrPlot$CondG, levels = c("BmC", "BmS"),labels = c("One Cursor","Two Cursors"))

head(cursorErrPlot)

A grouped_df: 6 × 5

Grp	Subj	CondG	meanIDE	meanPVDE
<fct>	<chr>	<fct>	<dbl>	<dbl>
YNC	Subj001	One Cursor	-3.1503930	-0.3632224
YNC	Subj002	One Cursor	-8.3391708	-7.3970882
YNC	Subj003	One Cursor	-1.1538569	-0.6782084
YNC	Subj004	One Cursor	-5.9896416	-6.2597010
YNC	Subj005	One Cursor	-0.2557512	-1.1746078
YNC	Subj006	One Cursor	-4.3837268	-2.0762970

options(repr.plot.width=10, repr.plot.height=6.5)
comp = list(c("YNC","RHP"),c("YNC","LHP"),c("LHP","RHP"))

meanIDE = ggplot(data = cursorErrPlot,aes(x = Grp, y = meanIDE,shape=Grp)) +
geom_half_boxplot(size = 1,show.legend = FALSE) + geom_half_point_panel(size = 3,show.legend = FALSE) +
xlab("")+ ylab("cursor error (˚)") + 
coord_capped_cart(ylim = c(-20,20),bottom = "both", left = "both") +
stat_compare_means(comparisons = comp,label.y = c(12,15,18), na.rm = TRUE,label = "p.signif",
                   size=7,face="bold",bracket.size = 1,tip.length = 0.0) +  #facet_grid(~CondG)+
theme_pubclean() + thm

meanPVDE = ggplot(data = cursorErrPlot,aes(x = Grp, y = meanPVDE,shape=Grp)) +
geom_half_boxplot(show.legend = FALSE) + xlab("")+
coord_capped_cart(ylim = c(-12,12),bottom = "both", left = "both") +
stat_compare_means(comparisons = comp,label.y = c(5,7,9), na.rm = TRUE,label = "p.signif",
                   hide.ns = FALSE,vjust = 0, 
                    symnum.args=list(cutpoints = c(0.001, 0.01, 0.05,Inf),symbols = c("***", "**", "ns")),
                   size=8,face="bold",bracket.size = 1,tip.length = 0.0) +  
geom_half_point_panel(size = 3,show.legend = FALSE) + theme_pubclean() + thm
# ggsave(file="S3.svg", plot=meanIDE, width=10, height=6.5)


meanIDE | meanPVDE

Non-Paretic Left Hand Error

data  %>% 
filter(CondG==c("BmC")) %>% 
filter(Hand=="left")  %>%
filter(Grp!="LHP") %>% 
select(Grp,Subj,Hand,CondG,SubNo,Cond,Dist,Hand,Trial,IDE,PVDE) %>% 
group_by(Grp,Subj,Hand) %>%    
mutate(meanIDE = mean(IDE),meanPVDE = mean(PVDE)) %>% 
distinct(Subj,Grp,Hand,CondG,meanIDE,meanPVDE) %>% 
{.->>LhandErrPlot}
LhandErrPlot$Grp = factor(LhandErrPlot$Grp, levels = c("LHP", "YNC", "RHP"))

head(LhandErrPlot)

A grouped_df: 6 × 6

Grp	Subj	Hand	CondG	meanIDE	meanPVDE
<fct>	<chr>	<chr>	<chr>	<dbl>	<dbl>
YNC	Subj001	left	BmC	-7.348676	-11.3099125
YNC	Subj002	left	BmC	-22.905913	-21.5431340
YNC	Subj003	left	BmC	11.654869	2.6080192
YNC	Subj004	left	BmC	-8.591172	-15.7508540
YNC	Subj005	left	BmC	1.882509	-0.8127948
YNC	Subj006	left	BmC	-5.524778	-9.9725552

options(repr.plot.width=8, repr.plot.height=6.5)
comp = list(c("YNC","RHP"))

LIDE = ggplot(data = LhandErrPlot,aes(x = Grp, y = meanIDE,shape=Grp)) +
geom_half_boxplot(col="#110176",fill="#110176",alpha = .3,size = 1,width = 0.6,show.legend = FALSE) + 
geom_half_point_panel(col="#110176",size = 3,show.legend = FALSE)+
geom_hline(yintercept = 0,col="black",lty="dashed",lwd=1) + 
xlab("")+ ylab("error at peak acc (˚)") + 
coord_capped_cart(ylim = c(-25,20),bottom = "both", left = "both") +
stat_compare_means(comparisons = comp,label.y = c(15), na.rm = TRUE,
                   size=7,face="bold",bracket.size = 1,tip.length = 0.0) + 
theme_pubclean() + thm

LPVDE = ggplot(data = LhandErrPlot,aes(x = Grp, y = meanPVDE,shape=Grp)) +
geom_half_boxplot(col="#110176",fill="#110176",alpha = 0.3,size = 1,width = 0.6,show.legend = FALSE) + 
geom_half_point_panel(col="#110176",size = 3,show.legend = FALSE)+
geom_hline(yintercept = 0,col="black",lty="dashed",lwd=1) + 
xlab("")+ ylab("error at peak speed (˚)") + 
coord_capped_cart(ylim = c(-25,20),bottom = "both", left = "both") +
stat_compare_means(comparisons = comp,label.y = c(15), na.rm = TRUE,
                   size=7,face="bold",bracket.size = 1,tip.length = 0.0) + 
theme_pubclean() + thm

LErrFig = LIDE | LPVDE

# ggsave(file="S4.svg", plot=LErrFig, width=8, height=6.5)

LErrFig

---

For PNAS Revisions

---

Within trial gains, direction of causality

In individual correlation plots, switching x and y (theta_cursor with delta_theta)

options(repr.plot.width = 13.5, repr.plot.height = 4.5)

Ind_subjGL = 
ggplot(bmCData2 %>% filter(Grp=="LHP" & Subj=="Subj006"), aes(y = Adjustment, x = IDE_cursor,na.rm = TRUE,group = Hand)) +
       geom_point(aes(col = Hand),stroke = 1.5, size = 3.5, shape = 15, show.legend = FALSE,alpha = 0.3) + 
       geom_hline(yintercept = 0,size = 1,col="gray30",lty="dashed") + 
       geom_vline(xintercept = 0,size = 1,col="gray30",lty="dashed") + 
       stat_cor(aes(group = Hand,col=Hand,label =paste(..r.label.., cut(..p.., breaks = c(-Inf, 0.0001, 0.001, 0.01, 0.05, Inf),
                                                                        labels = c("'****'", "'***'", "'**'", "'*'", "'ns'")),
                                                       sep = "~")),method="spearman",label.x = -20, label.y = c(20,15), size=8,show.legend = FALSE) + 
       coord_capped_cart(xlim = c(-20,20),ylim = c(-20,20),bottom = 'both', left = 'both') + #facet_wrap(~Hand,ncol=2) +
       geom_smooth(aes(col=Hand),  method = 'lm', fill=NA,formula = y~x, lwd=3.5,show.legend = FALSE,fullrange=TRUE) +  
       scale_color_manual(values=c("#110176","#E93324")) +
       scale_fill_manual(values=c("#110176","#E93324")) + 
       scale_alpha_manual(values = c(0.7, 0.2)) +
       theme_pubclean() + thm +
       theme(panel.spacing.x=unit(-2, "lines")) + 
       theme(strip.text.x = element_text(size=0)) + theme(strip.background = element_blank(), strip.text = element_blank()) +
       ylab(expression(bold(Delta*theta))) + 
       xlab(expression(bold(theta)[Cursor]))

# ggsave(file='Ind_subjL.svg', plot=Ind_subjGL, width=4.5, height=4.5)

Ind_subjG = 
ggplot(bmCData2 %>% filter(SubNo==14), aes(y = Adjustment, x = IDE_cursor,na.rm = TRUE)) +
       geom_point(aes(col = Hand),stroke = 1.5, alpha = 0.3, size = 3.5, shape = 16, show.legend = FALSE) + 
       geom_hline(yintercept = 0,size = 1,col="gray30",lty="dashed") + 
       geom_vline(xintercept = 0,size = 1,col="gray30",lty="dashed") + 
       stat_cor(aes(group = Hand,col=Hand,label =paste(..r.label.., cut(..p.., breaks = c(-Inf, 0.0001, 0.001, 0.01, 0.05, Inf),
                                                                        labels = c("'***'", "'***'", "'**'", "'*'", "'ns'")),
                                                       sep = "~")),label.x = -20, label.y = c(20,15),size=7.5,show.legend = FALSE) + 
       coord_capped_cart(xlim = c(-20,20),ylim = c(-20,20), bottom = 'both', left = 'both') + #facet_wrap(~Hand,ncol=2) +
       geom_smooth(aes(col=Hand,fill=Hand), fill=NA,method = 'lm', formula = y~x, lwd=3.5,show.legend = FALSE,alpha=0.1,fullrange=TRUE) +  
       scale_color_manual(values=c("#110176","#E93324")) +
       scale_fill_manual(values=c("#110176","#E93324")) + theme_pubclean() + thm +
       theme(panel.spacing.x=unit(-2, "lines")) + 
       theme(strip.text.x = element_text(size=0)) + theme(strip.background = element_blank(), strip.text = element_blank()) +
       ylab(expression(bold(Delta*theta))) + 
       xlab(expression(bold(theta)[Cursor]))

# ggsave(file='Ind_subjG.svg', plot=Ind_subjG, width=4.5, height=4.5)

Ind_subjGR = 
ggplot(bmCData2 %>% filter(SubNo==22), aes(y = Adjustment, x = IDE_cursor,na.rm = TRUE,group = Hand)) +
       geom_point(aes(col = Hand),stroke = 1.5, size = 3.5, shape = 17, show.legend = FALSE,alpha = 0.3) + 
       geom_hline(yintercept = 0,size = 1,col="gray30",lty="dashed") + 
       geom_vline(xintercept = 0,size = 1,col="gray30",lty="dashed") + 
       stat_cor(aes(group = Hand,col=Hand,label =paste(..r.label.., cut(..p.., breaks = c(-Inf, 0.0001, 0.001, 0.01, 0.05, Inf),
                                                                        labels = c("'***'", "'***'", "'**'", "'*'", "'ns'")),
                                                       sep = "~")),digits=2,label.x = -12, label.y = c(20,15), size=7.5,show.legend = FALSE) + 
       coord_capped_cart(xlim = c(-20,20),ylim = c(-20,20),bottom = 'both', left = 'both') + #facet_wrap(~Hand,ncol=2) +
       geom_smooth(aes(col=Hand),  method = 'lm', fill=NA,formula = y~x, lwd=3.5,show.legend = FALSE,fullrange=TRUE) +  
       scale_color_manual(values=c("#110176","#E93324")) +
       scale_fill_manual(values=c("#110176","#E93324")) + 
       theme_pubclean() + thm +
       theme(panel.spacing.x=unit(-2, "lines")) + 
       theme(strip.text.x = element_text(size=0)) + theme(strip.background = element_blank(), strip.text = element_blank()) +
       ylab(expression(bold(Delta*theta))) + 
       xlab(expression(bold(theta)[Cursor]))

# ggsave(file='Ind_subjR.svg', plot=Ind_subjGR, width=4.5, height=4.5)

Ind_subjGALL = Ind_subjGL | Ind_subjG | Ind_subjGR

# ggsave(file='Ind_subjGALL.svg', plot=Ind_subjGALL, width=13.5, height=4.5)
Ind_subjGALL

In mixed effects models, switching x and y (adjustment with IDE_cursor).

• Adjustment varies by Hand and moderates the relationship with IDE_cursor, so 'Hand' included as FE and RE.
• Value of g does change from switching the x and y, but all main results still present.
• Interestingly, it seems more clear that the negative gains are present for both hands in YNC, i.e., both hands contribute to correction within trial, just Left > Right. This is what was suggested by R2 as well & Dr. G.
• In the right hemisphere stroke group, there is less compensation with both hands compared to YNC. This is why we refer to this as a "loss". Is there a better alternative word?

mixedModEstInter = suppressMessages(lmer(data = bmCData2, Adjustment ~   Hand + Grp + Grp:Hand:IDE_cursor +
    (1 + Hand + Hand:Grp:IDE_cursor| SubNo)))

coefs <- coef(summary(mixedModEstInter))
data.frame(coefs[, 1:2])

A data.frame: 10 × 2

	Estimate	Std..Error
	<dbl>	<dbl>
(Intercept)	-3.23579080	0.62795809
HandRight	5.61664172	0.86582681
GrpRHP	-0.06377817	0.67888420
GrpLHP	0.06018070	0.70039911
HandLeft:GrpYNC:IDE_cursor	-0.37724657	0.05020927
HandRight:GrpYNC:IDE_cursor	-0.23720165	0.07155430
HandLeft:GrpRHP:IDE_cursor	-0.09902340	0.04606562
HandRight:GrpRHP:IDE_cursor	-0.40718847	0.07530568
HandLeft:GrpLHP:IDE_cursor	-0.08474197	0.04442909
HandRight:GrpLHP:IDE_cursor	-0.18575119	0.08265325

Left = c(coef(mixedModEstInter)$SubNo[1:20,"HandLeft:GrpYNC:IDE_cursor"],
         coef(mixedModEstInter)$SubNo[21:32,"HandLeft:GrpRHP:IDE_cursor"],
         coef(mixedModEstInter)$SubNo[44:47,"HandLeft:GrpRHP:IDE_cursor"],         
         coef(mixedModEstInter)$SubNo[33:43,"HandLeft:GrpLHP:IDE_cursor"],
         coef(mixedModEstInter)$SubNo[48:50,"HandLeft:GrpLHP:IDE_cursor"])

Right = c(coef(mixedModEstInter)$SubNo[1:20,"HandRight:GrpYNC:IDE_cursor"],
         coef(mixedModEstInter)$SubNo[21:32,"HandRight:GrpRHP:IDE_cursor"],
         coef(mixedModEstInter)$SubNo[44:47,"HandRight:GrpRHP:IDE_cursor"],         
         coef(mixedModEstInter)$SubNo[33:43,"HandRight:GrpLHP:IDE_cursor"],
         coef(mixedModEstInter)$SubNo[48:50,"HandRight:GrpLHP:IDE_cursor"])


Grp = c(matrix('YNC',20),matrix('RHP',15),matrix('LHP',15)) #,

Subj = c(1:length(Left))
Slopes = data.frame(Subj,Grp,Left,Right)

# Pivot data table for analysis & plotting
Slopes  %>% 
pivot_longer(cols = c(Left,Right),names_to = "Hand",values_to = "g")  %>% 
{.->>MMEstGains}
MMEstGains$Grp = factor(MMEstGains$Grp, levels=c('LHP','YNC','RHP'))
head(MMEstGains)

A tibble: 6 × 4

Subj	Grp	Hand	g
<int>	<fct>	<chr>	<dbl>
1	YNC	Left	-0.45767043
1	YNC	Right	-0.36023260
2	YNC	Left	-0.25721714
2	YNC	Right	-0.06565796
3	YNC	Left	-0.74347454
3	YNC	Right	-0.14888867

options(repr.plot.width = 11, repr.plot.height = 7)
options(warn = -1)

rs1 = ggplot(data = MMEstGains, aes(x = Grp, y = g, shape = Grp, col = Hand)) + stat_summary(mapping = NULL,
    data = NULL, fun.data = mean_se, alpha = 0.8, geom = "pointrange", position = position_dodge(width = 0.5),
    size = 1.5, lwd = 3, show.legend = FALSE) + geom_half_point_panel(aes(col = Hand),
    size = 3, alpha = 0.3, position = position_dodge(width = 0.5), show.legend = FALSE) +
    geom_line(aes(group = Hand, col = Hand), size = 2, fun.data = mean_se, alpha = 1,
        position = position_dodge(width = 0.5), stat = "summary", show.legend = FALSE) +
    geom_hline(yintercept = 0, linetype = "dashed", size = 1) + coord_capped_cart(ylim = c(-0.75,
    0.75), bottom = "both", left = "both") + xlab("") + ylab("gain") + annotate_npc("paretic left",
    x = unit(c(0.08), "npc"), y = unit(c(0.38), "npc"), rot = 90, gp = gpar(fontsize = 20,
        fontface = 3, col = "#110176")) + annotate_npc("paretic right", x = unit(c(0.92),
    "npc"), y = unit(c(0.28), "npc"), rot = 90, gp = gpar(fontsize = 20, fontface = 3,
    col = "#E93324")) + scale_color_manual(values = c("#110176", "#E93324")) + scale_fill_manual(values = c("#110176",
    "#E93324")) + scale_shape_manual(values = c(15, 16, 17)) + theme_pubclean() +
    thm + theme(axis.line.x = element_line(colour = "white"), axis.text.x = element_blank(),
    axis.ticks.x = element_blank(), legend.position = "none")


Fig1 = gr1 + rs1

# ggsave(file = "Fig1_gain.svg", plot = Fig1, width = 11, height = 7)

Fig1
options(warn = 0)

Robust LMER to compare g between groups and hands

Comparing g between YNC & LHP

# SET REFERENCE
MMEstGains$Hand <- factor(MMEstGains$Hand, ordered = FALSE)
MMEstGains$Hand <- relevel(MMEstGains$Hand, "Right")
MMEstGains$Grp <- relevel(MMEstGains$Grp, "LHP")

# RUN ROBUST MODEL
robust.model1L = rlmer(g ~ Hand + Grp + Hand:Grp + (1 | Subj), MMEstGains %>% filter(Grp != "RHP"), doFit = TRUE)

#EMMEANS
summary(emmeans(robust.model1L, specs = pairwise ~ Hand | Grp)$contrasts)
summary(emmeans(robust.model1L, specs = pairwise ~ Grp | Hand)$contrasts)

A summary_emm: 2 × 7

	contrast	Grp	estimate	SE	df	z.ratio	p.value
	<fct>	<fct>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	Right - Left	LHP	-0.1351433	0.04148137	Inf	-3.257929	1.122285e-03
2	Right - Left	YNC	0.1528362	0.03592392	Inf	4.254441	2.095717e-05

A summary_emm: 2 × 7

	contrast	Hand	estimate	SE	df	z.ratio	p.value
	<fct>	<fct>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	LHP - YNC	Right	0.006636981	0.05703829	Inf	0.1163601	9.073672e-01
2	LHP - YNC	Left	0.294616512	0.05703829	Inf	5.1652410	2.401294e-07

Comparing g between YNC & RHP

# SET REFERENCE
MMEstGains$Hand <- factor(MMEstGains$Hand, ordered = FALSE)
MMEstGains$Hand <- relevel(MMEstGains$Hand, "Right")
MMEstGains$Grp <- relevel(MMEstGains$Grp, "RHP")

# RUN ROBUST MODEL
robust.model1R = rlmer(g ~ Hand + Grp + Hand:Grp + (1 | Subj), MMEstGains %>% filter(Grp != "LHP"), doFit = TRUE)

#EMMEANS
summary(emmeans(robust.model1R, specs = pairwise ~ Hand | Grp)$contrasts)
summary(emmeans(robust.model1R, specs = pairwise ~ Grp | Hand)$contrasts)

A summary_emm: 2 × 7

	contrast	Grp	estimate	SE	df	z.ratio	p.value
	<fct>	<fct>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	Right - Left	RHP	-0.2846553	0.04341018	Inf	-6.557339	5.477632e-11
2	Right - Left	YNC	0.1545537	0.03759432	Inf	4.111092	3.937915e-05

A summary_emm: 2 × 7

	contrast	Hand	estimate	SE	df	z.ratio	p.value
	<fct>	<fct>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
1	RHP - YNC	Right	-0.1366932	0.05097017	Inf	-2.681828	7.322102e-03
2	RHP - YNC	Left	0.3025158	0.05097017	Inf	5.935153	2.935711e-09

Left hand errors result from an inability to compensate for inertial properties and interjoint dynamics as shown in previous work. Our data in neurotypical controls support this.

data %>% 
filter(CondG==c("BmC","BmS")) %>% 
filter(Hand!="cursor") %>% 
filter(Grp=="YNC") %>% 
mutate(Trial = as.numeric(Trial)) %>% 
select(Subj,SubNo,CondG,Dist,Hand,Trial,IDE,PVDE,XPosPV,YPosPV,CURV) %>% 
group_by(Subj,CondG,Hand) %>% 
summarize(mnIDE = median(IDE),
          mnPVDE = median(PVDE),
          mnCurv = median(CURV),.groups="drop") %>% 
{.->>LErrorsDat}

head(LErrorsDat)

LErrorsDat$CondG = factor(LErrorsDat$CondG, levels = c("BmC", "BmS"),labels = c("One Cursor","Two Cursors"))

A tibble: 6 × 6

Subj	CondG	Hand	mnIDE	mnPVDE	mnCurv
<chr>	<chr>	<chr>	<dbl>	<dbl>	<dbl>
Subj001	BmC	left	-6.678415	-10.420480	0.169935
Subj001	BmC	right	-0.110695	4.759570	0.126945
Subj001	BmS	left	-6.329610	-10.853010	0.169910
Subj001	BmS	right	-4.097490	3.834125	0.119385
Subj002	BmC	left	-23.366285	-22.371495	0.294570
Subj002	BmC	right	10.431790	10.018570	0.123570

Initial directional errors and curvature of the left hand from our neurotypical data

comp = list(c("left","right"))

theta_peakAcc = 
ggplot(LErrorsDat, aes(y=(mnIDE),x=Hand,col=Hand)) + 
  geom_half_dotplot(aes(col=Hand,fill=Hand),binwidth=1.5,alpha = 0.7,show.legend = FALSE)+ 
  geom_half_boxplot(aes(col=Hand,fill=Hand),alpha = 0.2,size = 1, show.legend = FALSE)+ 
coord_capped_cart(ylim= c(-30, 20),bottom = 'both', left='both') + facet_grid(~CondG) +
geom_hline(yintercept = 0,lwd=1.5,lty="solid",col="black") + ylab("initial directional error") + xlab("") +
scale_color_manual(values = c("#110176", "#E93324")) + scale_fill_manual(values = c("#110176","#E93324")) +
stat_compare_means(comparisons = comp,label.y = c(15), na.rm = TRUE,
                   size=7,face="bold",bracket.size = 1,tip.length = 0.0) + 
theme_pubclean() + theme(axis.line.x = element_line(colour = "white"), axis.text.x = element_blank(), axis.ticks.x = element_blank()) +
 ylab(expression(bold(theta)[italic(peak~acc)])) + ggtitle("A. Directional error is larger for the left hand") +
thm

curvature = 
ggplot(LErrorsDat, aes(y=(mnCurv),x=Hand,col=Hand)) + 
  geom_half_dotplot(aes(col=Hand,fill=Hand),binwidth=0.015,alpha = 0.7,show.legend = FALSE)+ 
  geom_half_boxplot(aes(col=Hand,fill=Hand),alpha = 0.2,size = 1, show.legend = FALSE)+ 
coord_capped_cart(ylim= c(0, 0.4),bottom = 'both', left='both') + facet_grid(~CondG) +
scale_color_manual(values = c("#110176", "#E93324")) + scale_fill_manual(values = c("#110176","#E93324")) +
stat_compare_means(comparisons = comp,label.y = c(0.34), na.rm = TRUE,
                   size=7,face="bold",bracket.size = 1,tip.length = 0.0) + 
theme_pubclean() + theme(axis.line.x = element_line(colour = "white"), axis.text.x = element_blank(), axis.ticks.x = element_blank()) +
 ylab("path curvature (au)") + xlab("") + ggtitle("B. Hand path is more curved for the left hand") + thm

options(repr.plot.width=8, repr.plot.height=11)
newSupp1 = theta_peakAcc / curvature

# ggsave(file = "S2.svg", plot = newSupp1, width = 8, height = 11)
newSupp1

Time between moments of peak acc and peak speed and its comparison between groups

# colnames(data)[1] ="Grp"
data$Grp = factor(data$Grp, levels=c('YNC','RHP','LHP'))


data %>% 
filter(CondG==c("BmC")) %>% 
mutate(Adj = PVDE - IDE,FFDur = (TPV-PPT)*1000) %>%
filter(Hand!="cursor") %>% 
mutate(Trial = as.numeric(Trial)) %>% 
select(Grp,SubNo,Cond,Dist,Hand,Trial,IDE,PPT,Adj,FFDur) %>% 
group_by(Grp,SubNo,Hand) %>% 
summarize(mnAdj = median(Adj),
          mnFFDur = median(FFDur),.groups="drop") %>% 
{.->>AdjFF}

head(AdjFF)

AdjFF %>% 
group_by(Grp) %>% 
summarize(GrpMeanFF = median(mnFFDur), GrpSEFF = parameters::standard_error(mnFFDur))

A tibble: 6 × 5

Grp	SubNo	Hand	mnAdj	mnFFDur
<fct>	<int>	<chr>	<dbl>	<dbl>
YNC	1	left	-4.336570	51.72138
YNC	1	right	6.063695	60.34000
YNC	2	left	1.280545	51.72069
YNC	2	right	-0.330880	60.34000
YNC	3	left	-6.816130	43.10138
YNC	3	right	7.161920	43.10069

A tibble: 3 × 3

Grp	GrpMeanFF	GrpSEFF
<fct>	<dbl>	<dbl>
YNC	51.72069	1.547207
RHP	131.46621	24.583954
LHP	142.24276	22.948625

# Define function to calculate IQR at given quantiles
iqr = function(z, lower = 0.25, upper = 0.75) {
  data.frame(
    y = median(z),
    ymin = quantile(z, lower),
    ymax = quantile(z, upper)
  )
}

iqr(AdjFF$mnFFDur[AdjFF$Grp=="RHP"])
range(AdjFF$mnFFDur[AdjFF$Grp=="RHP"])

A data.frame: 1 × 3

	y	ymin	ymax
	<dbl>	<dbl>	<dbl>
25%	131.4662	61.42573	167.0239

1. -301.724138
2. 254.3127585

options(repr.plot.width=6, repr.plot.height=6)
my_comparisons = list(c("YNC","LHP"),c("YNC","RHP"),c("LHP","RHP"))
options(warn=-1)
AdjFFDurFig = 
ggplot(aes(x = Grp,y =(mnFFDur),shape=Grp),data =AdjFF) + 
stat_summary(fill="gray",geom="bar",width=.75, alpha = 0.8,fun = median,show.legend = FALSE) + 
stat_summary(fun.data = iqr, fun.args = list(lower = 0.25, upper = 0.75),  lwd=2, geom = "errorbar",width=0.2, show.legend = FALSE) + 
geom_point(aes(col=Hand,shape=Grp),na.rm=TRUE, alpha = 0.5,position = position_jitterdodge(jitter.width = 0.3,jitter.height = 0.2,dodge.width = 0.75),size=4,show.legend=FALSE) +
coord_capped_cart(ylim= c(0, 290),bottom = 'both', left='both') + 
scale_color_manual(values = c("#110176", "#E93324")) + scale_fill_manual(values = c("#110176","#E93324")) +
xlab("") + ylab("Early Phase\nDuration (ms)") +
ylab(expression(t[italic(peak~speed)]-t[italic(peak~acc)]~(ms))) +
stat_compare_means(show.legend=FALSE,vjust = 0.5,comparisons = my_comparisons,label.y=c(200,245,270), na.rm = TRUE,
                   size=12,face="italic",bracket.size = 1,tip.length = 0.0, hide.ns = TRUE,
                  symnum.args=list(cutpoints = c(0.0001, 0.001, 0.05,Inf),symbols = c("***", "**", "ns"))) +
theme_pubclean() + thm + theme(legend.position="none")
AdjFFDurFig
# ggsave(file = "AdjFFDurFig.svg", plot = AdjFFDurFig, width = 6, height = 6)

Figure 1E & F (+Supp Stroke): one-cursor vs two-cursor mean-centered in YNC, LHP and RHP

incl = colnames(data)[- (1:8)]
data  %>% 
filter(CondG==c("BmC","BmS")) %>% 

unite(col = "group", Grp,Subj,Cond,Trial) %>% 
filter(!group %in% rem2) %>% 
separate(col = group, into = c("Grp","Subj","Cond","Trial"),sep = "_") %>% 
pivot_wider(id_cols = NULL, names_from= Hand,values_from = all_of(incl)) %>% 
{.->>newData}
tail(newData,5)

newData$Grp = factor(newData$Grp, levels=c('YNC','RHP','LHP')) 
colnames(newData)

A tibble: 5 × 79

Grp	Subj	Cond	Trial	SubNo	CondG	Dist	CT_left	CT_right	CT_cursor	⋯	YPosPV_cursor	XPosOff_left	XPosOff_right	XPosOff_cursor	YPosOff_left	YPosOff_right	YPosOff_cursor	XDistPV_left	XDistPV_right	XDistPV_cursor
<chr>	<chr>	<chr>	<chr>	<int>	<chr>	<int>	<dbl>	<dbl>	<dbl>	⋯	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>
LHP	Subj016	BmC15	11	51	BmC	15	0.923	0.923	0.923	⋯	1.119468	0.9256264	1.071626	0.9993287	1.074570	1.156962	1.127890	-0.019821821	-0.013992917	-0.08600599
LHP	Subj016	BmC15	13	51	BmC	15	0.885	0.885	0.885	⋯	1.105135	0.9346889	1.072009	1.0027585	1.124245	1.119674	1.114799	-0.013083113	-0.015627742	-0.08395930
LHP	Subj016	BmC15	15	51	BmC	15	0.995	0.995	0.995	⋯	1.080734	0.9273329	1.065919	0.9965527	1.085324	1.099006	1.092654	-0.016961871	-0.020339184	-0.08830897
LHP	Subj016	BmC15	17	51	BmC	15	0.928	0.928	0.928	⋯	1.067673	0.9462309	1.081473	1.0117921	1.078281	1.097319	1.074236	-0.001889574	-0.005672671	-0.07449025
LHP	Subj016	BmC15	19	51	BmC	15	0.923	0.923	0.923	⋯	1.111223	0.9338825	1.068811	1.0009789	1.121591	1.124981	1.128809	-0.008924704	-0.016251533	-0.08224652

1. 'Grp'
2. 'Subj'
3. 'Cond'
4. 'Trial'
5. 'SubNo'
6. 'CondG'
7. 'Dist'
8. 'CT_left'
9. 'CT_right'
10. 'CT_cursor'
11. 'RT_left'
12. 'RT_right'
13. 'RT_cursor'
14. 'PPT_left'
15. 'PPT_right'
16. 'PPT_cursor'
17. 'TPV_left'
18. 'TPV_right'
19. 'TPV_cursor'
20. 'MT_left'
21. 'MT_right'
22. 'MT_cursor'
23. 'PV_left'
24. 'PV_right'
25. 'PV_cursor'
26. 'EE_left'
27. 'EE_right'
28. 'EE_cursor'
29. 'VE_left'
30. 'VE_right'
31. 'VE_cursor'
32. 'IDE_left'
33. 'IDE_right'
34. 'IDE_cursor'
35. 'PRDE_left'
36. 'PRDE_right'
37. 'PRDE_cursor'
38. 'PVDE_left'
39. 'PVDE_right'
40. 'PVDE_cursor'
41. 'FDE_left'
42. 'FDE_right'
43. 'FDE_cursor'
44. 'FPE_left'
45. 'FPE_right'
46. 'FPE_cursor'
47. 'IDIST_left'
48. 'IDIST_right'
49. 'IDIST_cursor'
50. 'PVDIST_left'
51. 'PVDIST_right'
52. 'PVDIST_cursor'
53. 'FDIST_left'
54. 'FDIST_right'
55. 'FDIST_cursor'
56. 'CURV_left'
57. 'CURV_right'
58. 'CURV_cursor'
59. 'OnAxErr_left'
60. 'OnAxErr_right'
61. 'OnAxErr_cursor'
62. 'OffAxErr_left'
63. 'OffAxErr_right'
64. 'OffAxErr_cursor'
65. 'XPosPV_left'
66. 'XPosPV_right'
67. 'XPosPV_cursor'
68. 'YPosPV_left'
69. 'YPosPV_right'
70. 'YPosPV_cursor'
71. 'XPosOff_left'
72. 'XPosOff_right'
73. 'XPosOff_cursor'
74. 'YPosOff_left'
75. 'YPosOff_right'
76. 'YPosOff_cursor'
77. 'XDistPV_left'
78. 'XDistPV_right'
79. 'XDistPV_cursor'

## FOR SPATIAL COOUPLING FIGURE

rXPos_bmC = vector()
rXPos_bmS = vector()

SubjID = vector()

for (i in newData$SubNo)
    {rXPos_bmC[i] = cor.test(newData$XDistPV_left[newData$SubNo==i & newData$CondG=="BmC"],
                             newData$XDistPV_right[newData$SubNo==i & newData$CondG=="BmC"],
                             method = "spearman")$ estimate
     rXPos_bmS[i] = cor.test(newData$XDistPV_left[newData$SubNo==i & newData$CondG=="BmS"],
                             newData$XDistPV_right[newData$SubNo==i & newData$CondG=="BmS"],
                             method = "spearman")$ estimate

     SubjID[i] = toString(i)
    }


rXPos = data.frame(rXPos_bmC,rXPos_bmS)


colnames(rXPos)[1] <- "BmC"
colnames(rXPos)[2] <- "BmS"


j = c(matrix('YNC',20),matrix('RHP',15),matrix('LHP', 16)) #,

rData = data.frame(SubjID,data.frame(j), rXPos)

colnames(rData)[2] = "Grp"

rData  %>% 
pivot_longer(cols = c("BmC","BmS"),"Cond",values_to = "corr")  %>% 

{.->> rData2}

rData2$Grp = factor(rData2$Grp, levels=c('YNC','RHP','LHP'))

head(rData2,4)

A tibble: 4 × 4

SubjID	Grp	Cond	corr
<chr>	<fct>	<chr>	<dbl>
1	YNC	BmC	-0.9157895
1	YNC	BmS	-0.6571429
2	YNC	BmC	-0.9639098
2	YNC	BmS	-0.3669173

display_markdown('### EXP vs CTRL condition in YNC')
options(repr.plot.width=12, repr.plot.height=6)

glance(t.test(rData2$corr[rData2$Grp=="YNC"&rData2$Cond=="BmS"],
       rData2$corr[rData2$Grp=="YNC"&rData2$Cond=="BmC"],paired = TRUE))
C1 = 
ggplot(aes(x = (XDistPV_right)*100, y=(XDistPV_left)*100, col=CondG),
           data = newData %>% filter(Grp=="YNC")) + 
    geom_point(size=2.5,alpha = 0.7) + stat_ellipse(lwd=2) + 
    coord_capped_cart(xlim = c(-6,6),ylim = c(-6,6), bottom = 'both', left='both') + 
    geom_abline(slope = -1,lwd=1) + 
    ylab("") + xlab("") + 
    theme_pubclean() + thm +  facet_wrap(~Grp) +
    scale_color_manual(values=c("#503010","#B5833F"),name = "",labels = c("One cursor","Two cursors")) + 
    scale_fill_manual(values=c("#503010","#B5833F"))

D1 = 
ggplot(rData2 %>% filter(Grp=="YNC"), aes(x=corr)) + 
    geom_density(aes(y=..scaled..,colour=Cond, fill=Cond),alpha = 0.5, size=2,show.legend = TRUE) + 
    geom_rug(aes(col = Cond),size = 1.4,length = unit(0.07, "npc"),outside = FALSE, sides = "b") + 
    geom_vline(xintercept = 0,col="black",size=1,linetype = "dashed")+ 
    coord_capped_cart(ylim = c(-0.1,1), xlim = c(-1, 1), bottom = 'both', left='both') + 
    ylab("density") + theme_pubclean() + thm + facet_wrap(~Grp) +
    scale_color_manual(values=c("#503010","#B5833F"),name = "",labels = c("One cursor","Two cursors")) + 
    scale_fill_manual(values=c("#503010","#B5833F"),name="", labels = c("One cursor","Two cursors"))

YNC = C1 + D1 + plot_annotation(tag_levels = 'A') & theme(plot.tag = element_text(size = 22))
YNC
# ggsave(file = "YNC_1_2_corr.svg", plot = YNC, width = 12, height = 12)

EXP vs CTRL condition in YNC

A tibble: 1 × 8

estimate	statistic	p.value	parameter	conf.low	conf.high	method	alternative
<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<chr>	<chr>
0.4640602	6.370511	4.125443e-06	19	0.3115937	0.6165266	Paired t-test	two.sided

display_markdown('## EXP vs CTRL condition in RHP')
options(repr.plot.width=12, repr.plot.height=6)

glance(t.test(rData2$corr[rData2$Grp=="RHP"&rData2$Cond=="BmS"],
       rData2$corr[rData2$Grp=="RHP"&rData2$Cond=="BmC"],paired = TRUE))

C2 = 
ggplot(aes(x = (XDistPV_right)*100, y=(XDistPV_left)*100, col=CondG),
           data = newData %>% filter(Grp=="RHP")) + 
    geom_point(size=2.5,alpha = 0.7) + stat_ellipse(lwd=2) + 
    coord_capped_cart(xlim = c(-6,6),ylim = c(-6,6), bottom = 'both', left='both') + 
    geom_abline(slope = -1,lwd=1) + 
    ylab("") + xlab("") + 
    theme_pubclean() + thm +  facet_wrap(~Grp) +
    scale_color_manual(values=c("#503010","#B5833F"),name = "",labels = c("One cursor","Two cursors")) + 
    scale_fill_manual(values=c("#503010","#B5833F"))

D2 = 
ggplot(rData2 %>% filter(Grp=="RHP"), aes(x=corr)) + 
    geom_density(aes(y=..scaled..,colour=Cond, fill=Cond),alpha = 0.5, size=2,show.legend = TRUE) + 
    geom_rug(aes(col = Cond),size = 1.4,length = unit(0.07, "npc"),outside = FALSE, sides = "b") + 
    geom_vline(xintercept = 0,col="black",size=1,linetype = "dashed")+ 
    coord_capped_cart(ylim = c(-0.1,1), xlim = c(-1, 1), bottom = 'both', left='both') + 
    ylab("density") + theme_pubclean() + thm + facet_wrap(~Grp) +
    scale_color_manual(values=c("#503010","#B5833F"),name = "",labels = c("One cursor","Two cursors")) + 
    scale_fill_manual(values=c("#503010","#B5833F"),name="", labels = c("One cursor","Two cursors"))

RHP = C2 + D2 + plot_annotation(tag_levels = 'A') & theme(plot.tag = element_text(size = 22))
RHP

EXP vs CTRL condition in RHP

A tibble: 1 × 8

estimate	statistic	p.value	parameter	conf.low	conf.high	method	alternative
<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<chr>	<chr>
0.5504428	4.421112	0.0005807458	14	0.2834099	0.8174757	Paired t-test	two.sided

display_markdown('## EXP vs CTRL condition in LHP')
options(repr.plot.width=12, repr.plot.height=6)

glance(t.test(rData2$corr[rData2$Grp=="LHP"&rData2$Cond=="BmS"],
       rData2$corr[rData2$Grp=="LHP"&rData2$Cond=="BmC"],paired = TRUE))

C3 = 
ggplot(aes(x = (XDistPV_right)*100, y=(XDistPV_left)*100, col=CondG),
           data = newData %>% filter(Grp=="LHP")) + 
    geom_point(size=2.5,alpha = 0.7) + stat_ellipse(lwd=2) + 
    coord_capped_cart(xlim = c(-6,6),ylim = c(-6,6), bottom = 'both', left='both') + 
    geom_abline(slope = -1,lwd=1) + 
    ylab("") + xlab("") + 
    theme_pubclean() + thm +  facet_wrap(~Grp) +
    scale_color_manual(values=c("#503010","#B5833F"),name = "",labels = c("One cursor","Two cursors")) + 
    scale_fill_manual(values=c("#503010","#B5833F"))

D3 = 
ggplot(rData2 %>% filter(Grp=="LHP"), aes(x=corr)) + 
    geom_density(aes(y=..scaled..,colour=Cond, fill=Cond),alpha = 0.5, size=2,show.legend = TRUE) + 
    geom_rug(aes(col = Cond),size = 1.4,length = unit(0.07, "npc"),outside = FALSE, sides = "b") + 
    geom_vline(xintercept = 0,col="black",size=1,linetype = "dashed")+ 
    coord_capped_cart(ylim = c(-0.1,1), xlim = c(-1, 1), bottom = 'both', left='both') + 
    ylab("density") + theme_pubclean() + thm + facet_wrap(~Grp) +
    scale_color_manual(values=c("#503010","#B5833F"),name = "",labels = c("One cursor","Two cursors")) + 
    scale_fill_manual(values=c("#503010","#B5833F"),name="", labels = c("One cursor","Two cursors"))

LHP = C3 + D3 + plot_annotation(tag_levels = 'A') & theme(plot.tag = element_text(size = 22))
LHP

EXP vs CTRL condition in LHP

A tibble: 1 × 8

estimate	statistic	p.value	parameter	conf.low	conf.high	method	alternative
<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<dbl>	<chr>	<chr>
0.5837261	5.480136	8.106093e-05	14	0.3552704	0.8121817	Paired t-test	two.sided

---

END OF ANALYSIS

---

display_markdown("### Packages in use: ")
search()

display_markdown("### R version & citation: ")
citation()

display_markdown("### Specific package to cite: EMMEANS")
citation("emmeans")

Packages in use:

1. '.GlobalEnv'
2. 'package:ggplotify'
3. 'package:influence.ME'
4. 'package:svglite'
5. 'package:broom'
6. 'package:emmeans'
7. 'package:table1'
8. 'package:sjstats'
9. 'package:sjPlot'
10. 'package:nlme'
11. 'package:grid'
12. 'package:ggExtra'
13. 'package:robustlmm'
14. 'package:gghalves'
15. 'package:lme4'
16. 'package:Matrix'
17. 'package:lemon'
18. 'package:tidyr'
19. 'package:dplyr'
20. 'package:patchwork'
21. 'package:ggpubr'
22. 'package:ggplot2'
23. 'package:repr'
24. 'package:IRdisplay'
25. 'package:jsonlite'
26. 'package:formatR'
27. 'jupyter:irkernel'
28. 'package:stats'
29. 'package:graphics'
30. 'package:grDevices'
31. 'package:utils'
32. 'package:datasets'
33. 'package:methods'
34. 'Autoloads'
35. 'package:base'

R version & citation:

To cite R in publications use:

  R Core Team (2020). R: A language and environment for statistical
  computing. R Foundation for Statistical Computing, Vienna, Austria.
  URL https://www.R-project.org/.

A BibTeX entry for LaTeX users is

  @Manual{,
    title = {R: A Language and Environment for Statistical Computing},
    author = {{R Core Team}},
    organization = {R Foundation for Statistical Computing},
    address = {Vienna, Austria},
    year = {2020},
    url = {https://www.R-project.org/},
  }

We have invested a lot of time and effort in creating R, please cite it
when using it for data analysis. See also ‘citation("pkgname")’ for
citing R packages.

Specific package to cite: EMMEANS

To cite package ‘emmeans’ in publications use:

  Russell V. Lenth (2021). emmeans: Estimated Marginal Means, aka
  Least-Squares Means. R package version 1.7.0.
  https://CRAN.R-project.org/package=emmeans

A BibTeX entry for LaTeX users is

  @Manual{,
    title = {emmeans: Estimated Marginal Means, aka Least-Squares Means},
    author = {Russell V. Lenth},
    year = {2021},
    note = {R package version 1.7.0},
    url = {https://CRAN.R-project.org/package=emmeans},
  }